METSÄNTUTKIMUSLAITOKSEN TIEDONANTOJA 764, 2000 FINNISH FOREST RESEARCH INSTITUTE, RESEARCH PAPERS 764,2000 Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland Timo Kärki JOENSUU RESEARCH STATION METSÄNTUTKIMUSLAITOKSEN TIEDONANTOJA 764 GREY ALDER (Alnus incana) AS A RAW MATERIAL FOR MECHANICAL WOOD PROCESSING IN FINLAND Timo Kärki ACADEMIC DISSERTATION To be presented with the permission of the Faculty of Forestry of the University of Joensuu, for public criticism in Auditorium C 2 of the University, Yliopistokatu 2, Joensuu, on April 14, at 12 o'clock noon. JOENSUUN TUTKIMUSASEMA 2000 2 Kärki, T. Supervisors: Pertti Harstela, Professor Forest Technology University of Joensuu Erkki Verkasalo, Professor Wood Science and Technology Finnish Forest Research Institute, Joensuu Research Station Reviewers: Bohumil Kucera, Professor Department of Forest Economics, Operations and Wood Science Norwegian Forest Research Institute Harri Mäkinen, Ph.D. Growth and Yield, Wood Science Finnish Forest Research Institute, Vantaa Research Centre Opponent: Marketta Sipi, Professor Department of Forest Resource Management University of Helsinki Publisher: The Finnish Forest Research Institute, Joensuu Research Station, P.0.80x 68, FIN-80101 Joensuu, Finland. Accepted to be published by Matti Kärkkäinen, Research Director. Distribution: The Finnish Forest Research Institute, Library, P.0.80x 18, FIN -01301 Vantaa, Finland. Tel. +358-9-8570 5580, Fax. +358-9-8570 5582, email: library@metla.fi. Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland 3 ABSTRACT Kärki, T. 2000. Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland. D. Sc. thesis. The Finnish Forest Research Institute, Research Papers 764. 48 p. ISBN 951-40-1722-6, ISSN 0358-4283. In the Savo-Karelia area, the attitudes of sawmillers towards using grey alder as raw material were found to be positive. It was also determined in Study 11, with regard to tree species that 10 % of all furniture in Central European markets was made of alder. This means that alder was the fourth most popular tree species for furniture after pine, beech and spruce. The construction of different quality models indicated that in grey alder stands there is much unexplained variation. In the case of the tree height model, tree diameter was the strongest explanatory variable. The model concerning height to the lowest living branch showed similar results. In the case of the lowest dead branch, only the height to the first living branch could be used as a predictor. In all height models, the largest variation was the residual variation between trees in a plot. The technical quality of grey alder logs and sawn timber varies considerably. Grey alder stems have knots from the butt to the crown, and the length of the knot free section at the butt is short. The boundary between a dry-knot section and a fresh-knot section is not as obvious as, for example, with softwoods. There have not previously been volume tables for grey alder, so the volume table for birch with bark was chosen as a point of comparison for the whole tree-volume table. It was concluded that, for extreme sizes of grey alder stems, the volume estimates for grey alder differ 10-11 % from the estimates calculated with the volume tables of birch. The theoretically calculated stumpage price for grey alder logs was 158 FIM/m 3 . This result is a computational price that is based on average estimates. However, this price gives an indication of the price level suitable for grey alder logs in Finland. It was also shown in the furniture manufacturing calculation that the raw-material price for alder is not yet the limiting factor. The price limit for good quality alder sawn timber could be 3 000 FEM/m 3 . Keywords: Alnus incana, grading, value, quality classification, forest product markets. Author's address: Finnish Forest Research Institute, Joensuu Research Station, P.O. Box 68, 80101 Joensuu, Finland. Fax +358-13-251 4111, email Timo. Karki @ metla.fi . 4 Kärki. T. Esipuhe Kiinnostukseni lepän tutkimukseen alkoi vuosina 1993-94, kun kiersin puuta jalostavien yrittäjien luona Savo-Karjalassa keräämässä aineistoa pro gradu - työtäni varten. Tällöin selvisi yrittäjien ilmeinen mielenkiinto harmaaleppään raaka-aineena, mutta keskusteluissa tuotiin usein ilmi esteet näiden puulajien käytön kehittämiselle: sahauksen ja kuivauksen ongelmat, puun laadun ennustettavuuden ongelmat sekä laatu- ja mittakriteerien puute. Lisäksi lepän markkinointia haluttiin kehittää, sillä yrittäjien tuntemuksien mukaan paikallisia markkinoita oli, mutta mikä oli käyttö muualla, suuressa maailmassa ? Mihin leppää yleensä käytetään ? Tällöin hahmottui mielessäni kokonaisuus, jossa lähdettäisiin esimerkiksi huonekaluteollisuuden vaatimuksia hyväksikäyttäen etenemään kohti tuotannon alkupäätä eli metsää tavoitteena laskea näin lepän sahauksen kannattavuus. Haasteelliseksi tehtävän teki väite siitä, että suomalaisista lepikoista ei ole muuhun käyttötarkoitukseen kuin hellapuuksi, useimpien mielestä ei siihenkään. Lisäksi kuulin usein mielipiteitä, että leppä on kansainvälisestikin katsottuna "nolla" puulaji, jolla ei ole arvoa taloudellisesti katsottuna muuallakaan maailmassa. Näillä perusteilla ja ajatuksilla aloin aktiivisesti hakea syksyllä 1995 rahoitusta tutkimukseni aloittamiseksi. Sain työhöni henkistä sekä ideallista tukea professori Pertti Harstelalta, jota tahdon tässä yhteydessä erityisesti kiittää luovasta sekä idearikkaasta ohjaustyöstä tutkimuksen aikana. Lisäksi tahdon kiittää professori Erkki Verkasaloa, joka on valaissut minulle lehtipuututkimuksen metodologiaa sekä professori Matti Kärkkäistä, joka oli ideoimassa varsinkin markkinatutkimusta. Edellä mainittujen lisäksi tahdon kiittää niitä lukuisia henkilöitä, jotka ovat edesauttaneet väitöskirjani syntymistä. Taloudellisesti työtäni ovat tukeneet Joensuun yliopisto, Suomen Kulttuurirahasto, Suomen Akatemia, Maa- ja metsätalousministeriö, Metsämiesten Säätiö sekä Metsätieteellinen Seura. Rahoittajille parhaat kiitokset. Joensuussa helmikuun kirkkaina talvipäivinä vuonna 2000 Timo Kärki Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland 5 CONTENTS LIST OF ORIGINAL PAPERS 6 1 INTRODUCTION 7 2 WOOD PROCUREMENT, PROCESSING AND MARKETS FOR GREY ALDER-A LITERATURE REVIEW 8 2.1 Alder in Finland 8 2.2 The physical and mechanical properties of alder timber 11 2.3 Wood procurement of alder 13 2.4 Markets for alder timber 15 2.5 Profitability of sawmilling and furniture manufacturing 20 3 AIM OF THE STUDY 22 4 MATERIAL AND METHODS 23 5 RESULTS 25 5.1 Markets for alder timber (I, II) 25 5.2 Tree and log quality of grey alder (111, IV) 26 5.3 Volume tables for grey alder trees and logs (V) 30 5.4 A theoretical calculation for processing value of grey alder (11, 111, IV, V)... 32 6 DISCUSSION 37 7 CONCLUSIONS 42 REFERENCES 43 ORIGINAL PAPERS I - V 6 Kärki, T. LIST OF ORIGINAL PAPERS This dissertation is a summary of the following papers, which are referred to in the text by Roman numerals I - V. I Kärki, T. 1997. Sahauskelpoisen erikoispuun laatuvaatimukset ja käyttö Savo-Karjalan alueella. Folia Forestalia 1/1997: 37-48. II Kärki, T. 1999. Species, furniture type, and market factors influencing furniture sales in southern Germany. Accepted to Forest Products Journal 12 th January 2000. Will be published in Volume 50 (4). 14 p. ID Kärki, T„ Maltamo, M. & Eerikäinen, K. 1999. Diameter distribution, stem volume and stem quality models for grey alder (.Alnus incana) in eastern Finland. Accepted to New Forests 7 th December 1999. In print. 22 P- IV Kärki, T. 1999. Predicting the value of grey alder logs (Alnus incana) based on external quality. Silva Fennica 33(1): 13-23. V Kärki, T., Eerikäinen, K., Heinonen, J. & Korhonen, K.T. 1999. Harmaalepän (Alnus incana) tilavuustaulukot. Folia Forestalia 1/1999: 39- 49. NB: These articles should be referred to according to the bibliographic information given above. No reference should be made to the reprint in this thesis. The author is responsible for the following in the joint papers: In papers 111 and V the author was having main responsibility in planning of the sampling, measuring and analysing, and writing the results. Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland 7 1 INTRODUCTION Improved use of grey alder has been under discussion in Finland for the past 10-15 years. Recently, there has been further interest in using grey alder as raw material for the mechanical wood industry. The fact is that the potential of hardwoods will increase in future; for example, Siitonen (1990) calculated that in 1999-2019 the harvesting possibilities for hardwoods (excluding birch) will be doubled because of the increased growth and volume of hardwoods. In these calculations the effects of changes in silviculture (biodiversity and preferring hardwoods) are not considered. Although the matter has interested many people and institutions, the „case of grey alder" has been little studied. In Finland, research on grey alder began in the 1930's with a review of its utilization for mechanical processing as well as for pulping and papermaking (Routala & Sihtola 1934). The fibre-morphological and pulping characteristics of grey alder were studied by Bruun & Slungaard (1957) and Alestalo & Hentola (1967). The results showed that, compared with other Finnish tree species, grey alder is short-fibred. Compared with birch, sulphate pulps of grey alder have a short beating time and good tension and bursting strength, but poorer tearing strength. Other technical properties such as basic density, proportion of bark and dry matter content of grey alder were studied by Lehtonen et ai. (1978), Mali (1980), Farmer (1981) and Björklund & Ferm (1982). The wood of alder is homogeneous and easy to work, making it suitable for various useful objects and many special purposes, e.g. making music instruments (Salmi 1977 and Kärki 1997). The yield of grey alder has been studied by Miettinen (1933) and Mäkinen (1984). Saarsalmi et al. (1985, 1991, 1992) studied the biomass production of grey alder in short rotations and found that growth of young grey alder is better than that of pine, spruce, birch or aspen. However, the growth of grey alder begins to slow down and the occurrence of decay increases at the age of 25-30 years (Miettinen 1933, Tikka 1954, Ilvessalo & Ilvessalo 1975 and Vuokila 1977). The profitability of growing grey alder depends on its suitability as a raw material for industry. Even though the possibilities and markets for grey alder timber have been studied so little, a literature review is presented in chapters 2.1 - 2.5 in order to give an idea of the wood procurement, use, processing and markets of grey alder. The wood procurement is reviewed by studying new possible trade forms, timber exchange and terminal trade, use and processing by looking into the current state of the use of alder timber in Finland and markets by clarifying the potential markets for grey alder timber in Germany, which is the most important export market for products of the Finnish mechanical forest industry. 8 Kärki, T. 2 WOOD PROCUREMENT, PROCESSING AND MARKETS FOR GREY ALDER - A LITERATURE REVIEW 2.1 Alder in Finland In Finland, grey alder is considered a special tree species and the timber of grey alder as special timber. Both definitions are also used in this dissertation. According to Kärki (1995), special tree species can be defined as species which either grow outside their natural area of distribution or cover a small proportion of the forest land / annual growth or are different in structure and character from the main species growing in a particular area. Special timber is defined as timber that has a particular end-use or dimensions or is little used. Alder is distributed throughout Finland, with the exception of the northeasternmost part of the country. There are two alder species, grey alder (.Alnus incana) and black alder (Alnus glutinosa). Grey alder can often be found in eastern Finland on formerly cultivated land, to which grey alder as a rapidly growing pioneer tree species has spread. The proportion of grey alder of the total number of trees is considerable; but in terms of stand management, grey alder has been cut down and the volume of grey alder compared to the number of stems is marginal (Figures 2 and 3). Black alder grows mainly in southern Finland as naturally regenerated and planted forests (Holmäsen 1991) and is typically a species for wet biotope-sites like shore forests, although grey alder prefers dryer areas than black alder does. Alder species are known as tree species that improve the quality of soil (Huss-Danell & Lundmark 1987). Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland 9 Figure 1. Natural area of distribution for grey alder in Europe (Holmåsen 1991). Grey alder, which typically occurs in semiboreal and boreal forests, has spread out further northward to the limit of forest growth than black alder has. From Finland the area of distribution for grey alder continues to eastward the forest zone of Russia, to the Caucasus Mountains, Siberia, the Kamtchatka peninsula and Japan. From Finland southward, grey alder can be found in much of Central Europe with the exception of Britain, Denmark, northern Germany, Holland, Belgium, western France, the Pyrenean peninsula and the southern part of the Balkan peninsula. Familiar forms of grey alder can also be found in the northern part of North America. (Holmäsen 1991) The proportions of aspen, black alder and grey alder in terms of stem number and volume in southern Finland (15 southest Forestry board districts) are shown in Figures 2 and 3. In order to compare the proportions of different potential special tree species in Finland, the proportions of aspen are also shown. These figures are based on the results of the Eighth National Forest Inventory (VMI8). 10 Kärki. T. Figure 2. Proportions of aspen, black alder and grey alder based on stem number in southern Finland. On the basis of stem number, the proportion of alder and aspen is greatest in Itä-Häme (23.1 percent). The proportion of grey alder is largest in the forest districts of Itä-Häme, South Savo, Itä-Savo, North Karelia and North Savo. On average, in Southern Finland the proportion of grey alder according to stem num ber, is 9.6 percent. Black alder occurs in low proportions; in Aland, Helsinki and Itä-Savo, however, the proportion is over 1 percent. Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland 11 Figure 3. Proportion of aspen, black alder and grey alder of the total tree volume in southern Finland. The proportion of aspen, grey and black alder calculated on the basis of tree vol ume is greatest in Aland (5.8 %) (Figure 3). The next three are quite equal, Itä- Häme (4.5 %), North Savo (4.2 %) and Helsinki (4.2%). In the provinces of Aland and Helsinki, the proportion of black alder is the greatest; in the other districts aspen and grey alder dominate. In conclusion, it can be said that the proportion of grey alder, determined according to both stem number and basal area, is greatest on former burnt-over areas in eastern Finland. When total volume is considered, the amount of grey alder is spread out evenly throughout southern Finland. 2.2 Physical and mechanical properties of alder timber There are three common alder species in the world, all of which are widely used in mechanical wood processing. As mentioned previously, grey and black alder grow in Finland, while red alder (Alnus rubra) is wide-spread in North America. Despite the fact that there are three different species of alder, alder timber is commonly sold as one species. In Table 1, the most important physical and mechanical properties for mechanical wood processing of different alder species are compared. Cherry is used as a comparison tree species, while alder is frequently mentioned as substitute for cherry in furniture manufacturing (Grosser 1985 and Kohonen 1997). 12 Kärki. T. Table 1. Physical and mechanical properties of grey alder, black alder, red alder and cherry (Hakkila 1970, Salmi 1977, Farmer 1981, Björklund & Ferm 1982, Grönros et al. 1995 and Wagenfuhr 1996). When the physical properties of alder species are compared, grey alder is the lightest of the three alder species, and shrinkage is also greatest in grey alder, followed closely by black alder. On the other hand, the mechanical properties of grey alder seem to be noticeably better than those for red alder. The mechanical properties of black alder are virtually the same as for grey alder, and it is men tioned in some references that they behave equally in finishing (Salmi 1977, Mali Characteristics Grey alder Black alder Red alder Cherry Alnus incana Alnus glutinosa Alnus rubra Prunus serotina Physical properties Density (kg/m3) - oven-dry 361 490 430 530 - dry-air dry 430 530 460 580 - green 600 850 737 900 Shrinkage (%) - longitudinal 0.35 0.35 - 0.2 - radial 4.5 4.0 4.0 4.0 - tangential 9.5 9.7 7.0 7.0 - volumetric 14.1 14.2 12.6 11.5 Mechanical properties Bending strength (MPa/m 2 ) 88.3 83.4 44.6 54.9 Compression strength (MPa/m2) 44.1 39.2 20.3 24.3 Hardness (HJ -L, N/mm2 ) 43 43 20 29 Tensile strength (MPa/m2) - - 2.6 3.9 Shearing strength (MPa/m2) 4.9 4.4 5.3 7.7 Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland 13 1980 and Grönros 1995). The wood of grey alder is reddish white and slightly lighter than that of black alder. Sapwood and heartwood are usually indistinguish able, as is also true for black alder. According to Mali (1980) and Grönros et ai. (1995), the behaviour of grey alder wood e.g. in drying, machining and finishing is easy to control. The wood of freshly felled black alder is reported to have a warm light brown colour, which changes rapidly to deep bright orange. The colour of the processed wood gradually fades to dull nut brown. The wood of black alder is reported to dry well and fairly rapidly, and degradation during seasoning is mini mal. Bending strength and resistance to shock loads are reported to be low, and stiffness is rated as very low. The timber of Alnus glutinosa is very difficult to distinguish from that of Alnus incana (Laidlaw 1960 and Farmer 1981). The heartwood of red alder, which is indistinct from the sapwood, is whitish when freshly cut but upon exposure to light changes to reddish-brown. Red alder has been described as "pleasing" in appearance, with grain that is typically straight and subdued. Although the colour has often been described as delicate, the wood lacks any outstanding figure. The wood is reported to be easy to dry, and seasons fairly rapidly with little degradation. Specially developed kiln schedules, careful handling of logs and freshly milled sawn timber are reported to be essential for obtaining and maintaining uniformity of colour during drying. Varying colora tion ranging from yellow to deep red, sometimes mottled, is caused by oxidation of the extractives present in the wood. The bending strength of the species is consid ered to be medium, being much weaker than white oak or teak in air-dry condition (about 12 percent moisture content). Red alder timber is not hard and can be dented and marred easily. Therefore, it is not recommended for uses that require strength. (Kline 1987, USDA 1987 and Niemiec et al. 1995) 2.3 Wood procurement of alder timber Wood procurement of grey alder nowadays is mainly as standing sale, sale at de livered price or cash transaction. There has been an interest in developing new business forums, like timber exchange and terminal trade, for trading of grey alder timber. The first models for timber exchange have been provincial exchanges (run by forest societies associations and the Finnish Forest Research Institute), so called information services, to which the person who wants to sell special timber announces the timber for sale and from which the buyer can for a small fee buy information about timber for sale. The trade and price are negotiated directly be tween seller and buyer. One new procurement possibility is wood procurement by a forest service enterpriser, in which an independent enterpriser buys the wood himself or sells timber directly to client and himself takes care of the logging and transport. The idea of a timber exchange is quite simple. The sellers and buyers in stock give their offers to an exchange system through the internet where the trades are made according to certain rules. In order to work properly, there must be 14 Kärki. T. enough sellers and buyers in the exchange. The problem in establishing a timber exchange is that there are no recognized quality and dimension classifications that could be used for special timber. It has also been unclear how to take into account in the price the time of delivery and terms of payment. The level of logging and transport costs has also been unclear. According to Kallio & Salo (1992), there are two possible forms for timber exchange: automatic exchange and real-time ex change. The trades are decided in the automatic exchange according to bids that have been announced by a certain time and the trades are then made immediately. The real-time exchange works differently; these trades are made after the exchange period so that all interested parties can make trades as if they were at an auction (Kuula et al. 1992). The partners in timber exchange can be forest owners, firms or even inde pendent forest-service enterprisers. Organisations that might run timber exchanges could be, for example, stock exchanges, trading houses, service companies or separate joint enterprises (Vepsäläinen & Kuula 1992). According to this idea, individual sellers could transact business through different advisors (like forestry associations). Another new possibility for trading grey alder timber is terminal trade. The typical problem in trading grey alder timber is that the quantities of timber for sale are small. In this case, there is a need to collect the small amounts together many times in order to obtain larger trade amounts if there are no suitable road-side landings to which small amounts can be transported to wait for long-distance transport. According to Siikaluoma (1991), terminal trade means sale at delivery in which the timber quantities are measured and/or accepted at so-called "village terminals". Terminal sites could be areas that have no other particular use, like former gravel pits. Terminal trade could also be practised at sawmills. In terminal trade, timber is delivered from roadside landings to terminals with tractors or lorries by timber sellers or buyers. This kind of trades are, ac cording to Siikaluoma, unprofitable unless the transport from forest to terminal is short and the quantities very large. Investment in expensive forest equipment is seen as a hindrance to profitability. On the other hand, Siikaluoma (1991) sees the possibilities of terminal trade in trading special timber. Cost gains could be reached in the first place in actual measuring and in changeovers from one stock to another. This principle is well adapted to special timber measurement, while part of the timber is rarely measured or can be measured with the measuring equipment on a harvester (Mäkelä 1986). Kärki (1995) calculated the wood procurement costs for timber assort ments of grey alder with a spreadsheet calculation of Metsäteho (Oijala et al. 1994). The average wood procurement costs for grey alder logs were calculated to be 99.85 FIM/m3 and for grey alder pulpwood 142.76 FIM/m3. The calculation was made for a spruce thinning where grey alder timber was supposed to grow and be harvested after 30 years (manual felling, transport and administrative costs in cluded). On average, the wood procurement costs from a thinning (logs and pulp wood) with manual felling and transport were 114,70 FIM/m 3 (Oijala 1994), so the wood procurement of grey alder logs was 15 FIM/m 3 cheaper and wood procure Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland 15 ment of grey alder pulpwood 43 FIM/m 3 more expensive than for thinnings on average according to the example calculation. Kärki (1995) also calculated the procurement costs in a situation where grey alder would grow as a pure stand for final felling. In this case, the wood procurement costs were considerably smaller, being then equivalent to birch. 2.4 Markets for alder timber The use of alder ( Alnus glutinosa and A. incana) has increased rapidly in the past few years, among other things, in sauna-furnishing and manufacturing panels. Al der is also used in furniture, utility articles, ornaments and in the cellulose and fibreboard industry (Salmi 1978a,b) In the furniture industry alder is much utilized as a competitive alternative to walnut, cherry and mahogany (Grosser 1989). Grey and black alder are used for the same type of end-products. The amount of grey alder timber in the Finnish markets seems to be limited (Table 2). The special trees used most in Finland are aspen, alder, larch and curly birch. The amounts of other domestic hardwoods like oak are marginal. At the same time, in 1993 35 000 m 3 of hardwood timber (excluding birch) was imported to Finland. (Louna & Valkonen 1995). Table 2. Use of domestic special timber in Finland in 1995. Use as firewood is excluded. (Verkasalo 1993, Louna & Valkonen 1995, Kärki 1995, Hynynen & Viherä-Aarnio 1999) Tree species Used amount Aspen Logs: 10 000 - 15 000 m 3 Pulpwood: 150 000 - 200 000 m3 Alder Logs: 20 000 m3 Pulpwood: 30 000 m3 Curly birch Some hundred kilos Larch 1 000 - 2 000 m3 Oak 100 m3 Mountain ash < 100 m3 Maple < 100 m3 Linden < 100 m3 Ash < 100 m3 Elm < 100 m3 16 Kärki, T. It must be noted that the estimates of Louna & Valkonen (1995) for alder were calculated by using data in which only the larger firms (over 20 employees) were represented; from the smaller firms there was only a 10 % sample. This indicates that the importance of smaller firms is underestimated. A more realistic estimate for the use of grey alder in mechanical wood processing could be about 50 000 m 3 per year. This rough estimate is based on the supposition that most alder timber is processed mechanically by small saw mills as a supplement to softwood processing. Alder is counted as one of the „trend tree species" in the German markets, which are the most important export markets for the products of Finnish mechani cal forest industries. In German markets, the use of alder timber has increased rapidly during the last five years. In these markets, some domestic grey and black alder timber is also available, but most of the alder timber is imported from the United States and eastern and southern Europe. Import of alder from eastern Europe has increased significantly in the last few years. The import from these countries is very competitive in price, and most of this timber is imported directly to the processers, bypassing the timber importers (Kräftiger Anstieg ... 1998). However, in this study the import of hardwoods to Germany is examined only through import from the largest exporting country, the United States, because ac curate statistics on the import of alder from eastern and southern Europe are not available. The import of hardwood timber to Germany from the United States, divided into tree species and sortiments, in 1997 is presented in Table 3. Table 3. Import of hardwood timber to Germany from the United States divided into tree species and sortiments in 1997 (values in mill. FTM). (Handelsstrukturen für ... 1998) Tree Round Sawn Veneers Total Proportion species timber timber (%) Ash 11.34 11.34 15.99 38.67 3 Birch - 2.65 1.62 4.27 0 Cherry 69.71 6.37 63.99 140.07 12 Walnut - 1.13 - 1.13 0 (hickory & pecan) Maple 9.45 62.10 39.20 110.75 9 Red oak 5.02 3.35 59.51 67.88 6 Walnut 1.40 2.97 44.01 48.38 4 Red alder 0.97 335.88 - 336.85 28 White oak - 139.86 123.12 262.98 22 Yellow poplar 1.35 6.59 - 7.94 1 Other 19.44 16.69 162.65 198.78 16 Total 118.68 588.93 510.09 1217.7 100 Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland 17 When different hardwood species are considered, the proportion of im ported alder is greatest (28 %). After alder, the proportions of imported white oak (22 %) and cherry (12 %) are next and the group „other species" makes up 16 % of the import of hardwoods. It is interesting that alder imported from the United States is almost all sawn timber while, e.g. for white oak, the proportion sawn tim ber / veneers is equal in value. Cherry is imported mostly as round timber and ve neers, but small amounts of sawn timber are also imported. It seems that cherry is imported as unprocessed round timber and alder as slightly processed sawn timber; there is a possibility that cherry is sawn or manufactured into veneer, whereas al der is used as sawn timber for furniture and veneered by slicing. Not all alder is used for the furniture industry; it is also used for other end-products such as par quet. The import of hardwood timber from the United States to the countries of the European Union in 1996 and 1997 is presented in Figure 4. Figure 4. Import of hardwood sawn timber to the countries of the European Uni on from the United States in 1996 and 1997. (Handelsstrukturen für ... 1998) 18 Kärki, T. According to Figure 4, the import of hardwood timber to the European Union from the United States has increased in all countries from 1996 to 1997, also to the group which comprises the other members of the European Union, in cluding Finland. In the case of Germany and Italy, the import has increased above all due to the increased import of alder (Handelsstrukturen... 1998). There are no statistics available concerning the price of alder sawn timber per cubic meter imported to Germany from the United States, but a rough estimate (EXW delivery conditions, ex works) can be calculated using the information in Table 3 and Figure 4. From Table 3 we can see that of all hardwood sawn timber the value of alder sawn timber imported to Germany from the States is 57 %, the value of white oak 24 % and the value of maple 11%. These three most important species comprise 92 % (538 mill. FIM) of the whole value. The value of imported hardwood sawn timber (588.93 mill. FIM) divided by the number of imported cu bic metres (170 000 m 3) gives about 3 500 FIM / m 3. It can be said, according to tree species, that the average price per cubic metre for oak and maple is higher than that for alder. If we suppose that this is so, the average price per cubic metre for imported alder sawn timber is about 3 000 FIM / m 3. In 1996, the average price per cubic metre for beech imported to Finland from Germany was 3 500 FIM / m 3, which can be used as a reference price for alder (Außenhandel nach ... 1997). Other interesting reference prices for sawn timber are the import prices for hard wood and softwood imports to Germany from Finland in 1996: spruce 990 FIM / m 3, pine 1 790 FIM /m 3 and birch 2 470 FIM / m 3. The latter prices are delivered at frontier (DAF) prices, so, to be able to compare the cubic prices with the other before mentioned (Außenhandel nach ... 1997), we have to add some hundred FIM to the price of the alder sawn timber. The value of hardwood veneers imported to the countries of the European Union from the United States in 1996 and 1997 is presented in Figure 5. Grey alder ( Alnus incana) as a raw material for mechanical wood processing in Finland 19 Figure 5. Value of hardwood veneers imported to the European Union from the United States in 1996 and 1997. The values are in FTM according to EXW conditions. (Kräftiger Anstieg ... 1998) During 1996 and 1997, the import of veneers to Germany and Spain from the United States increased slightly, but the import to the United Kingdom decreased slightly. As seen in Table 3, according to value, the veneers are mainly white oak and cherry; the proportion of imported alder veneers is marginal. The prices for hardwood logs in German markets are presented in Figure 6. 20 Kärki, T. Figure 6. Hardwood log prices (B-quality) in the State Forests of Baden- Württemberg and Bayern 1990-98. (Jahresberichte der Baden- Württembergische und Bayerische Staatsforsten 1991-1998). The price of alder, including both black and grey alder, has been increas ing slightly in the last seven years, and the price is fixed to about 600 FTM/m 3 . The price of alder has followed that of beech very closely and was 20-30 % higher than the price of birch. 2.5 Profitability of sawmilling and furniture manufacturing No studies are available concerning the profitability of sawmilling and furniture manufacturing from grey alder timber. There are, however, some studies which can be used as a basis for calculating the value of grey alder for mechanical wood processing (Chapter 5.4). Kuukka (1980), Lönnberg & Vesikallio (1980), Ryti (1987), Heino (1988 a, b) and Pöyhönen (1991) investigated the profitability of sawing and processing of softwoods. Kuukka (1980), Heino (1988a,b) and Pöy hönen (1991) studied the profitability of individual sawmills. Heino (1988 a) came to the conclusion that the smaller the sawmill capacity, the more profitable the sawmill was. However, the range of annual production was as narrow as 40-50 000 m 3 to 100-110 000 m 3. Moreover, the change in the natural distribution area of tree species reduces the profitability of sawmilling. This is due to the rising costs of wood procurement. Heino's conclusion that the relationship between yield and grade had no effect on profitability is also interesting. As a major conclusion, he proposes some possible business strategies. One of them is a strategy where the sawmill Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland 21 - specializes in special wares - invests in marketing - invests in developed controlling system - invests in "flexible" equipment - controls the manufacturing process from the forest to the markets. The average cost structure of the Finnish sawmills in 1975-85 was also presented by Heino (1988 a). Of the total costs, the proportion of raw material was 68 %, labour 13 %, energy 3 %, maintenance 3 %, other variable costs 9 % and capital costs 12 %. According to these figures, the average deficit was 7 %. It was also found that the relative proportion of energy and maintenance from the total costs increased as the capacity of the sawmill grew. Heino (1988b) compared the profitability of sawmills and wood processing. In his report he mentions that the accuracy of bookkeeping at sawmills, if any, is poor. The raw material costs were 50-70 % (average 58 %) of the total costs at the sawmill. The productivity of capital was poor, even negative at sawmills, whereas the profit rate was 3.1% at the further processing mills. As a conclusion, there was a need to specialize in a particular product. Accordingly, no mill needs to make all types of products, but sawmillers should invest in individual sawing strategy, taking into account the raw material resources of the area, the competition and the market conditions. 22 Kärki. T. 3 AIM OF THE STUDY According to the literature reviewed, mechanical processing of grey alder has been studied very little. No studies are available, e.g. on the timber quality and stem volumes of grey alder. The main aim of this dissertation was to investigate the suitability of grey alder for mechanical wood processing in Finland. The aim di vides into following subaims: i) evaluating the markets for grey alder in Finland and Germany ii) evaluating the tree and saw log quality of natural born grey alder forests in Finland iii) calculating the volume and quality functions for grey alder logs iv) calculating the processing value for grey alder timber based on the articles presented in this dissertation Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland 23 4 MATERIAL AND METHODS The materials and methods of the various studies (I - V) are presented in more detail in each article. Therefore in this chapter the material and methods are out lined only generally. Four different sets of data were used in this study. In Study I, the empirical material was collected with a postal inquiry from firms that processed wood mechanically in the Savo-Karelia area. In Study n, the method used was an advertisement study. The material in that study consists of advertisements collected in southern Germany in the economic area of Munich. The data were collected from advertisements published in provincial and local newspapers and from the advertisements of furniture stores. Both of these first studies are descriptive, partly social research that charts the situation in practise. The materials in Studies 111 and V included 33 stands situated in the provinces of Northern Karelia and Northern Savo in eastern Finland. The stands were naturally regenerated, but at least part of them were managed. The material is measured data and thus differs considerably from the material in Studies I and EL The research material in Study IV comprised 229 grey alder saw logs from two pure stands of grey alder, both situated in Northern Savo, one in Maaninka and the other in Nilsiä. The trees were selected randomly, with the only limitation being that the logs had to be sawable (stem form). The methods used and reported in different articles are quite heterogene ous. The methods used and their influence on calculation of the processing value are shown in Table 4. 24 Kärki, T. Table 4. Methods used in the articles and the relevance for the calculation of the processing value. Different types of regression analysis were used in three articles. Cross tabulation was used in all five articles, and it was the only method in Article 11. In Article H, a Chi-Square Test was also attempted, but no significant differences were found. Thus, the test was omitted from the article. The main ideas of diffe rent methods, like principal component analysis and regression analysis, are clari fied in each article, so the different methods are not presented in detail here. Article I was the beginning of for the whole dissertation. It was found in Study I, that grey alder was the most interesting special tree species and should be investigated as a raw material for mechanical wood processing. The reason for this variable group of articles that includes aspects of marketing research, forest men suration and wood technology is that there were no real studies that could be used as a basis for calculation of the processing value. Volume models were lacking, as well as information about the saw log and sawn timber quality of grey alder. The possible processing value for grey alder timber is estimated with a theoretical calculation (Chapter 5.4). The information from Articles 11, 111, IV and V is used in different stages of the calculation as follows: 1. The structure of the example forest was calculated by using the diameter dis tribution, height and volume models from Articles 111 and V. Article Method Used in the calculation Article I Principal component analysis Cross-tabulation - Article II Cross-tabulation Price level for the end products Article III Weibull distribution, maximum Simulation of the example likelihood estimation forest Regression analysis Logistic regression analysis Cross-tabulation Article IV Regression analysis Sawn timber yield, grade Cross-tabulation distribution of sawn timber saw logs and Article V Regression analysis Simulation of the example Cross-tabulation forest Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland 25 2. The quality criteria and volume tables from Articles IV and V were used in estimating the saw log and industrial tree volumes. 3. The costs of logging and transport were calculated using the information pre sented by Oijala (1994), Huotari (1995) and Kärki (1995). 4. The grades and values of the logs / sawn timber were estimated using the information in Article IV. 5. The costs of sawmilling were calculated based on Heino (1988 a) and Pöy hönen (1991). 6. The costs of furniture manufacturing were calculated based on Lahti (1991), Aravuo (1994), Juva & Ahtorinne (1995) and Huonekalujen valmistus (1996). 7. The log stumpage prices and sawn timber prices were calculated according to the information given above. Saw log and sawn timber prices were compared with the statistics reviewed here and in other studies. 5 RESULTS 5.1 Markets for alder timber (I, II) In Study I, the current quality and dimension criteria for grey alder used by the sawmillers in eastern Finland and the current state of the use of grey alder was investigated. The sawmillers limited the minimum top-diameter criteria for grey alder logs to 13 cm, but they also wanted to use smaller dimensioned grey alder logs if the stem form was suitable for sawmilling (Table 6; I). The minimum length for grey alder logs was set at 2.5 m (Table 7; I). A little decay was allowed, and sound knots were accepted by half of the sawmillers. In the Savo-Karelia area, the attitudes of the sawmillers towards using grey alder as a raw material were found to be positive. The tree species of the greatest interest was grey alder. According to Study I, the interest in mechanical processing of special timber is strong; 70 % of the responding companies were interested in extending their wood processing to special tree species. The wood used by the companies studied was obtained as their own procurement or directly from forest owners. The sawmillers wanted to develop wood procurement on this basis also in future. Of the sawmillers (12 %), were interested in developing their wood procurement through timber exchange and 28 % through any system that would work effectively. A principal component analysis was also made in order to study factors affecting the willingness of the enterprisers to use grey alder as a raw material for mechanical wood processing (Table 9; I). Five factors were built in analysis, of which one describes the behaviour of the whole population of sawmillers studied and four of which can be described as new production possibility, wood procure ment, markets and development projects. Part of the enterprisers thought that mec hanical processing of grey alder is a new production possibility. This could mean that processing grey alder was considered to be a balancer and diversifier of the 26 Kärki, T. production as the economic trends vary. Wood procurement of grey alder was con sidered to be a problem when mechanical processing of grey alder was started or increased. The reasons for this are uncertainty about suitable quality and dimension criteria and the unstable level of the stumpage price for grey alder. An important component in the analysis was markets for grey alder timber. According to the questionnaire results, customers had ordered alder or aspen timber from every third enterpriser and some of them could not meet the demand. The fourth component in the principal component analysis was development projects, which indicates that enterprisers are interested in and also need training, counselling and consultation concerning grey alder. The aim of the Study II was to investigate the furniture markets in Central Europe as a case study, focusing on the economic area of Munich, to determine how these markets are divided according to tree species, furniture type and argumentation used in furniture marketing, i.e. how the products are sold. A further aim was to ascertain how much alder timber is used in the furniture business. According to the results of this study, 10 % of all wooden furniture was made of alder. This means that alder was the fourth most popular tree species after pine, beech and spruce (Table 2; H). The proportion of alder furniture was even higher (15 %) for plate-structured furniture, 10 % for partly solid-structured furniture and 5 % for solid-structured furniture. Softwood furniture from pine and spruce was sold as solid-structured (Table 5; II). In plate-structured furniture, in particular, the proportions of these species were marginal. All tropical hardwood furniture of teak, mahogany and acacia was solid-structured. Alder and oak made up equal proportions of every structural type of furniture (solid-structured, partly solid- and plate-structured), but beech furniture was mostly plate-structured and only occasionally solid-structured. On the whole, oak furniture was the most expensive, with alder following as a close second. Furniture from beech was noticeably less expensive when used as partly solid but significantly more expensive as plate-structured. Pine and spruce furniture was the least expensive when it was solid and partly solid-structured. Plate-structured maple and beech furniture was considerably more expensive than that made from alder or oak. 5.2 Tree and log quality of grey alder (III, IV) In Study 111, different models were constructed for tree and stand characteristics of grey alder. These models included models for diameter distribution, tree height, height to the first living and dead branches, taper curve and occurrence of rot. The models constructed can be applied in two different situations: tree diameters are either measured or not. Parameter models of fitted basal area diameter distributions of the Weibull distributions are presented in Table 4 (HI). Median diameter was used as explanatory variable for both parameters. Constructed models are very simple including only one explanatory variable. The degree of determination is high for parameter b but quite low in the case of parameter c. This result is similar to some Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland 27 previous studies concerning Weibull distribution (e.g. Rennolls et al. 1985, Maltamo 1997). The predicted diameter distributions using parameter models (Table 4; IV) were validated using /"-statistics. The validation indicated that in only in 1 of the 33 cases the predicted and true distributions disagreed. In the tree height model, tree diameter was also the strongest explanatory variable (Table 5; III). The model concerning height to the lowest living branch showed similar results (Table 6; III). Stand and tree diameter and especially different height characteristics were used in this model. In the case of the lowest dead branch, only height to the first living branch could be used as a predictor (Table 6). This means that the applicability of this model is restricted to situations were some additional measurements in addition to normal stand mean characteristics are made. In all height models, the largest variation was the residual variation among trees in a plot. This is natural because sample trees represent different, and also extreme, positions in diameter distribution. The residual variation between stands was large for models for tree height and the lowest dead branch but in the case of height to the first living branch it was minimal A taper polynomial was calculated using over bark diameters measured at relative heights. This was done after visual analysis of stem form, in which the data were sorted into one centimetre diameter classes, and average taper curves were calculated for each diameter class using the relative diameters at relative heights. The hypothetical assumption that relative diameters at relative heights - i.e. relative stem form - do not change much, was verified. In most cases, when the taper curve approach is polynomial, a polynomial function has been based on the Fibonacci series (e.g. Laasasenaho 1982). In this study the combination of lower polynomial terms with one logarithmic term gave the best results. For grey alder, the polynomial taper curve function is presented in Equation 5 (III) and the parameter estimates and statistics for the taper polynomial are presented in Table 7 (III). The basic taper polynomial (Equation 5; HI) gives the same average stem form for all trees of different sizes. However, each tree has its own shape. Because there are certain regularities in the diameter differences at relative tree heights (Laasasenaho 1982), an adjusted basic model for a given tree can be generated as a function of the diameter at breast height and the total height of the tree (f(d,h)). Differences between observed diameters and the basic taper polynomial (fb(x)) were calculated; and correction equations (cpli) were estimated for diameter differences at relative tree heights of 10, 40 and 70 % (Table 8; HI). After the correction equations are applied, a cubic interpolation polynomial (Press et al. 1992; p. 114) can be calculated to pass through the calibration points. One extra point is needed for the tree top at relative height of 100 %, where the correction is 0. The final adjustment of the taper polynomial (fc(x)) is achieved by adding the corresponding coefficients of basic taper equation and interpolation polynomial (fr(x)) together: fc(x) = fb(x) + fr(x). The reliability of the taper polynomial as a volume predictor was tested in Study V. The proportion of trees with rot at stump height seemed to vary irregularly 28 Kärki, T. as a function of age (Figure 2; EI). The proportion of rot has its minimum value among the age classes of 30-50 years. A more unexpected result was that rot was more common in young age classes than in older ones. This could be a consequence of the regeneration in grey alder stands, because grey alder forests usually regenerate by sprouting from older trees. If the older trees are rotten, this means that some of the younger sprouts will also rot. In older age classes (over 50 years), the proportion of rotten trees increases as a function of rotation time. The very smallest trees seemed to be healthy (diameter class 4 cm), but already in diameter class 6 cm the proportion of rotten trees was 50 % (Figure 3; III). Evidently, just at the beginning the proportion of rotten sprouts is small but later the amount of rot increases, e.g. as a consequence of harvesting damage (Norokorpi 1979). The distribution of stands according to proportion of rotten sample trees in the stand is presented in Table 9 (HI). Almost half of the stands had rot in less than 20 % of the sample trees. On the contrary, 10 % of the stands had rot in almost all sample trees. By using the data collected, logistic regression models were constructed for predicting the probability of rot in stems. Explanatory variables were sought by calculating correlations between variables describing rot and other tree characteristics. Two models were constructed; the first (i=l) was made for predicting the probability of rot at stump height and the second (i=2) for predicting the probability of rot anywhere in a stem. The logistic model for calculating the occurrence of rot in grey alder stems is presented in Formula 3 (EH) and the coefficients of Equations yi and y2 are presented in Table 10 (DI). It can be deduced from the form of Formula 3 that the larger the value of the exponent the greater is the probability of rot in an individual tree. Considering the effects of the variables in Equation yl, the thicker the bark the greater is the probability that the tree has rot. The height of the lowest dead branch was also an indicator of rot; the higher the lowest dead branches are the smaller is the probability of rot. As was also seen in Figure 2 (EI), the probability of rot decreases as age increases. In the case of Equation y2 , the effect of bark is similar to the previous case, but the effect of height of the lowest dead branch is opposite. To be able to predict the length of rot in alder stems, two linear regression models were constructed (Table 11; HI). The first was made for cases where the rot diameter at stump height was not known and the second for cases where it was known. The reliability of the model without rot diameter at stump height as an explanatory variable was not very high. According to this model, the thicker the bark at 5 % height the longer is the rot in the stem, and the higher the first dead branch the shorter is the rot in the stem. Basal area was also statistically significant in this model. If the rot diameter at stump height is known, the reliability of the model is considerably better. In this particular model, the effects of the height of the lowest dead branch and basal area are similar to those in the previous model. According to Study IV, there were fewer fresh knots in small-dimensioned grey alder butt and top logs than in larger grey alder logs. Especially in large top 29 Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland logs, there were considerably more fresh knots than in other types of logs. In all types of logs, fresh knots were concentrated to the section from 1.5 to 2.5 m (Figure 1; IV). Dry knots were concentrated at the base (0-0.5 m) of the logs and at a distance of 1.5-2.5 m (Figure 2; IV). There were almost twice as many dry knots in the top logs as in the butt logs. In large butt logs there were fewer rotten knots than in other types of logs. Rotten knots occurred most frequently in small dimensioned top logs and were concentrated in the section from 0.5 to 2.5 m (Figure 3; IV). Large top logs had more knots than other types of logs did. In large top logs, there were twice as many fresh knots as in small-dimensioned top logs. In general, top logs were more knotted than butt logs were, which was consistent with the process of self-pruning. Rotten knots were more common in top logs than in butt logs (Figure 4; IV). Average values for different grades show that the grades are not equally distributed throughout the log (Table 7; IV). Grade A, i.e. knot-free sawn timber was mainly at the base of the logs. In this respect grades B, C and reject differed from each other only slightly. In Table 8 (IV), the logs are divided into four grades. The limit values show that each grade occurred equally in logs. Overall, it seems that in different types of logs the different grades were at the same distances. In the butt logs, grade C began noticeably farther away from the butt than in top logs. It also seems that the worse the grade was the greater was the length of the grade. So a rejected piece of sawn wood was usually rejected totally. The best grade A, was shorter than the other grades, and it could be combined with poorer grades (normally with grades B and C). The reasons for grade reductions are shown in Table 9 (IV). Number of dry knots was the reason for reduction in sawn timber grade in half of the cases. This happens in every grade. Another very remarkable reduction was seen in quality rejected; discoloration was the reason for quality change in almost half of the cases. One aim of this study was to investigate how suitable the quality classifications of Keinänen and Tahvanainen (Tables 2 and 3; IV) are. According to the classification (Table 2; IV), only 65 % of the logs could be classified to A, B or C grades and 35 % of the logs belonged to the log grade Rejected (Table 10; IV). The amount of log grade A was modest; only 7.7 % of the logs could be classified to grade A. 35 % of the saw logs were included in log grade B and 21 % in grade C. When the results in Table 11 (IV) are compared to the log grades measured (Table 10; IV), the main difference is the small amount of Rejected sawn wood. The volume of grade A sawn wood is significantly greater than could be predicted from the log grading. In Figure 5 (IV) we can see the relation between grades of the logs and quality of the sawn wood. From the figure can be seen what kind of sawn wood could be obtained from different grades of logs (A, B, C, Rejected). We can also see the average distribution of sawn wood qualities. From the grade-A saw logs we reached over 50 % grade-A sawn wood. The proportion of rejected sawn wood was very small (6 %). The sawn wood quality in log grades B and C do not differ. In both log grades the distribution of 30 Kärki, T. sawn wood grades is similar. The log grade Rejected includes the least grade-A sawn wood, but the volume of grades B and Rejected sawn wood are equal to the better sawlog qualities (B and C). Average amount of grade-A sawn wood in the sawlogs was 36 %, for grade-B 33 %, for grade-C 20 % and for the Rejected grade 11 %. The correlation matrix between the main predictors is shown in Table 12 (IV). These 9 factors were used when the final models were built. The other factors, such as twist and cracks, were not significant, so the correlation matrix for them is not presented. The largest correlations were between FK2 and DKI and between FKI and FK2. These correlations were taken into consideration when the regression models were built. The correlations between the predictors used in the models are smaller than +- 0.252 (between FK3 and TD). The regression models used for predicting the value for grey alder logs based on external quality are shown in Table 13 (IV). The models were made in order to predict the value of different kinds of logs. There are separate models for butt and top logs. In the explanation of the models there are differences in interpretations. In Model 1 (IV) there were no significant external signs of quality that could be used in prediction; obviously, for the knot-free base section of a log the only external signs that can be used are dimensional (e.g. top diameter and length). In Model 2 (IV) all the coefficients are negative, which means that all knots (fresh and dry) decrease the value of the stem. This is a direct result of preferring knot-free timber as the most valuable. Dry knots decrease the value more than fresh knots do. Of the quality signs, the knottiness at the section from 1.5 m to 2.5 meters is the most significant predictor. The standard error is smaller in Model 2 (IV) than in Model 1 (IV). In Model 3 (IV) the coefficient of determination was larger than in Model 2 (IV). Top diameter of a log was included in Model 3 and the coefficient of determination improved slightly (37 %). The explanations for the coefficients in this model were logical; the top diameter increases the value of a log, and the dry and fresh knots decrease the value. As shown in Model 2 (IV), dry knots decrease the value more than fresh knots do. As in Model 2 (IV), the knotness from 1.5 m to 2.5 meters was the most significant predictor of value. Of these six models, the standard error for Model 3 (IV) was the smallest. Model 4 (IV) gives the log value only according to the number of knots. The coefficients of determination for different variables are logical in interpretation. Both coefficients are negative, which indicates that knot-free timber is the most valuable. Dry knots influence the log value more than fresh knots do. In Model 5 (IV) the top diameter was included, which greatly increased the significance of the model. Here the coefficient of determination was 57 %, and the standard error was also smaller than in Model 4. Residual plots for the two most useful Models (3 and 5; IV) are shown in the Figure 6 (IV). The residual plot for top logs (incl. top diameter) has the smallest standard error (14.91). Model 5 (IV) has a large standard error but the residual is still symmetrical. Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland 31 5.3 Volume tables for grey alder trees and logs (V) The volume tables for grey alder are presented in Tables 6 - 9 (V). The volumes were determined as stem volumes with bark (dm 3 ) to an accuracy of one decimal for stems smaller than 100 dm3 and to an accuracy of one dm3 for stems bigger than 100 dm 3 . A model was estimated for predicting the volumes without bark (Equation 10; V). The coefficient of determination for this model was 0.015 and the p-value for diameter was 0.035. The coefficient of determination was low, while the variation in bark thickness was high and the stem diameter did not have any significant influence on bark thickness (see Table 10). A model was also constructed for predicting the volume without bark at any height (Equation 11). Previously, volume tables for grey alder have not been available. That is why the volume table for birch with bark (Laasasenaho & Snellman 1983) was chosen as a point of comparison for the whole tree-volume table (Table 7; V). The volume estimates for grey alder were smaller than the corresponding volumes for birch when the trees had large diameters (do > 15 cm) but low heights (h < 10 m). Then the volumes for grey alder in larger diameter classes are at most 10 percent smaller than for birch of the same size. Considering tall but small-diameter trees, larger volumes are obtained by using the volume tables for grey alder compared with the volume tables for birch. The differences are 11 %at most. This difference decreased to 6 % when the length of the trees was 20 m and diameter at breast height 20 cm. Both these examples were taken from the extreme volumes, where the values can be considered to be extrapolated. According to these new volume tables for grey alder and the tables for birch, the volumes for grey alder showed similar volumes in the middle-size trees (d,j = 10 cm, h = 10 m). Due to the small amount of data in the diameter classes more than 20 cm, the usability of volume estimates for grey alder stems larger than 20 cm must be considered carefully (Table 2; V). 32 Kärki. T. 5.4 Theoretical calculation for the processing value of grey alder (II, III, IV, V) The example forest An example forest was calculated in order to calculate the amount of logs and pulpwood. The calculation was made by using the models presented in Articles LH and V. The diameter distribution was calculated by using the Weibull function (Table 4, ID). The tree-height mixed model used is presented in Table 5 of Article 111. The grading system of Keinänen and Tahvanainen (1995) was used for the logs (minimum length 2.1 m and minimum top diameter 13 cm). The calculation is presented in Table 5. Table 5. Calculation of the example forest (area 1 ha, basal area 28 m2/ha, Hdom 15 m). Diam. Height Freq. Freq. * Volume of Volume of Log-% Log volume Pulpwood class (m) basal one tree one class (dm 3 ) (dm 3 ) (cm) area (dm 3 ) (dm 3 ) 4.00 5.92 0.00 84.63 4.48 378.84 378.84 6.00 8.63 0.03 278.15 12.10 3364.34 3364.34 8.00 10.54 0.06 319.89 25.39 8120.75 8120.75 10.00 11.91 0.08 299.71 44.51 13339.26 13339.26 12.00 12.94 0.10 258.08 68.94 17791.04 17791.04 14.00 13.74 0.12 210.40 97.96 20611.00 34 7007.74 13603.26 16.00 14.37 0.12 163.92 131.07 21484.65 62 13320.48 8164.17 18.00 14.88 0.11 122.41 168.17 20586.03 76 15645.38 4940.65 20.00 15.31 0.10 87.68 209.74 18389.79 83 15263.52 3126.26 22.00 15.66 0.08 60.22 256.84 15468.39 88 13612.19 1856.21 24.00 15.97 0.06 39.64 311.29 12340.07 90 11106.06 1234.01 26.00 16.23 0.05 24.98 375.74 9387.57 94 8824.31 563.25 28.00 16.46 0.03 15.06 453.94 6838.57 94 6428.26 410.31 30.00 16.66 0.02 8.68 551.19 4786.02 94 4498.85 287.16 32.00 16.84 0.01 4.78 674.96 3226.32 94 3032.74 193.58 34.00 17.00 0.01 2.51 835.94 2099.28 94 1973.32 125.96 36.00 17.15 0.00 1.26 1049.65 1320.68 94 1241.44 79.24 38.00 17.27 0.00 0.60 1339.05 804.43 94 756.16 48.27 40.00 17.39 0.00 0.27 1738.65 474.93 94 446.44 28.50 42.00 17.50 0.00 0.12 2301.27 272.04 94 255.72 16.32 181083.99 103412.61 77671.38 Total volume (m 3 ) 181.08 -logs 103.41 - pulpwood 77.67 33 Grey alder ( Alnus irtcana) as a raw material for mechanical wood processing in Finland According to the models used, the volume of the simulated alder stand is 181.1 m 3 / ha. Of that volume, 103.4 m 3 are logs and 77.7 m 3 pulpwood. When calculating the round wood price at the sawmill, we must include the estimated selling price of the end product or the onward selling price of the timber in the calculation. An example for calculating the timber price at the sawmill was presented in the dissertation of Pöyhönen (1991). The method of calculating the timber price at the sawmill includes the following steps: 1. The price of the sawn timber at the delivery storage. 2. The sales costs, sales rewards, reclamation costs, shipping and forwarding costs, delivery costs and transport costs from the sawmill to the delivery stock are subtracted from the sawn timber price. 3. The sales proceeds of the by-products at the sawmill are added. 4. As a result the product price at the sawmill is obtained. 5. The variable costs in the production are subtracted: product stock costs, manufacturing and repairing costs including social security costs, energy costs, by-raw material costs, material costs and chemical costs. 6. Fixed costs of the saw mill are subtracted: administration wages including social security costs, rents, insurances, office costs and other social costs. 7. Capital costs are subtracted: rate costs, depreciations which correspond to operating time of the machines and taxes. 8. The profit of the enterpriser is subtracted. 9. The final result is the raw material expenses at the sawmill. When the raw material expenses (FIM/m 3) are divided by the utilization factor (m 3 round wood / m 3 sawn timber), the price of one cubic meter of round wood at the storage of the sawmill is obtained. The statistical values used were collected from studies of Lallukka (1986), Ryti (1987), Heino (1988a,b), Pöyhönen (1991), Louna & Valkonen (1995), Kärki (1997) and Metsätilastollinen vuosikirja (1998). The values do not describe any particular sawmill. The numbers are intended to show the relationship between different factors in a situation where the processing of the example products is profitable. Furthermore, the sawmill is intended only for primary processing only and is expected to be located somewhere in eastern Finland. The logging costs used in the calculation are presented in the study of Kärki (1995). Overhead costs of the wood procurement include rates, administration, management, fixed assets and social costs. The long distance transport costs are calculated by using lorry transport (90 km) and logging costs consist of felling and processing, forwarding, additional costs and social costs. The volumes of different sawn timber grades are calculated according to Figure 5 and Table 5 in Article IV and the volumes of different sawlog qualities according to Table 10 in Article IV. The logs are sawn unedged (i.e. by flat slicing) followed by subsequent drying and ripping. Sawn wood is edged after drying. The variable costs in sawmilling include the costs of log storage, 34 Kärki, T. sawmilling, the sawn timber yard, drying, steam, power, running, material, repairing and social costs. The fixed costs consist of office, rent, car, insurance, administration and social costs and the capital costs include depreciations, rates and taxes. Table 6. Calculation of the unit price of sawn timber at the sawmill and stumpage price of saw logs. VOLUMES PRICE VALUE m FIM/m 3 FIM VOLUME OF THE EXAMPLE FOREST 181 - saw log volume 103.40 - pulpwood and slash 77.60 100 7760 AMOUNT OF SAW ABLE LOGS - saw log volume 103.40 - grade A 7.96 - grade B 37.12 - grade C 21.82 - pulpwood, slash 36.50 100 3650 Incomes from pulpwood and slash 11410 AMOUNT OF SAWN TIMBER Amount of saw log to the sawmill, 66.90 qualities A. B and C - amount of sawn timber (yield 36.5 %) 24.42 - quality A 10.49 2000 20983 - quality B 9.31 1300 12097 - quality C 4.62 500 2310 - pulpwood, slash 14.38 50 719 - bark (10 %) 6.69 50 334 - sawdust (5 %) 3.34 50 167 36612 The average sawn timber price / m 3 1499 Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland 35 Table 7. Calculation of the stumpage price (FIM/m 3). The stumpage price according to the example calculations is 158 FIM/m 3 (Tables 6 and 7). The stumpage price can change considerably, e.g. according to dimension and grade of the logs, wood procurement costs, production costs and, last but not least, the market price of lumber. Costs of furniture manufacturing The costs of manufacturing furniture from grey alder are calculated in Table 8. This calculation is made in order to evaluate the reliability of the calculated stumpage price. The product and operating concept of the example factory used in the calculations is presented by Juva & Ahtorinne (1995). The model factory is a typical plant manufacturing wooden dining-room tables and chairs. The calculation made here adapts the concepts used in those calculations. Some key points of the operation concept are: THE STUMPAGE PRICE VALUE, FIM / m3 Price of sawn timber on average 1499 Sales and delivery costs 65 Price of the sawn timber at the sawmill 1434 Sales of by-products 100 Sales proceeds (net) 1534 Reduced: Production costs Variable costs 683 Fixed costs 80 Production costs in total 763 Capital costs 50 Profit of the enterpriser 30 Total 843 Left for raw material costs 691 Left for saw log in storage 253 THE STUMPAGE PRICE FOR SAW LOG WITH BARK Price at the sawmill 253 Reduced: Overhead costs of the wood procurement 10 Long-distance transport costs 40 Logging costs 45 TOTAL 95 STUMPAGE PRICE 158 36 Kärki. T. there is one basic furniture model in the production, which is slightly modified for different products products are gathered in the store or at the customer's home the main raw material is solid hardwood; tags and other materials are carefully standardized the timber is bought as kiln-dried sawn timber The cost distribution at the example factory is presented in Figure 7. Figure 7. Cost distribution at the example factory. The calculation for manufacturing wooden chairs from grey alder and the costs of transportation and trade in furniture marketing for German markets is presented in Table 8. The estimated need for raw material is 0.035 m 3 (chair), but 0.05 m 3 is used in the calculation (Juva & Ahtorinne 1995). The price of grey alder sawn timber in the calculation is 2 500 FEM/m3 . Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland 37 Table 8. Calculation for manufacturing wooden chairs from grey alder and the costs of transportation and trade in furniture marketing. 6 DISCUSSION The high percentage of interested companies in the questionnaire made in Study I, can be partly explained by the fact that the companies which were not interested in using special timber did not answer the inquiry as enthusiastically as those that were interested. If we suppose that all companies in Savo-Karelia which were interested in the processing of alder replied to the inquiry, the real estimate of those interested would be 18 % (answering percentage 0.25 x percentage of all companies interested in the inquiry 0.70). According to this calculation, at least every fifth wood-processing company in Savo-Karelia is interested in processing special timber, including alder. The real percentage of interest can probably be found between these two extreme values. Even though all types of furniture stores were included in the material of Study 11, the most important source of error was the subjective sampling of advertisements. Along with the geographical concentration of the sampled material (economic area of Munich), this permits only limited possibilities to apply the results elsewhere in Central Europe. Unfortunately, the furniture purchasers were not interviewed; interviews would have made possible a detailed comparison between advertising arguments and purchaser requirements. Nevertheless, the material gives an indication of furniture markets in southern Germany. When the MANUFACTURING COSTS FIM/chair Wood material 44 % 145 Other materials 5 % 17 Labour 13 % 43 Production costs 5 % 17 Marketing costs 4 % 13 Buildings 11 % 36 Machines 13 % 43 Other rates 2 % 7 Profit 3 % 10 PRICE OF THE PRODUCT AT THE FACTORY 330 TRADE Freight 50 FIM 380 Marginal of the wholesale and retail trade 30 % 522 VAT 16 % (Germany) 624 Price of the product 639 MARKETING PRICE OF A WOODEN CHAIR FROM GREY ALDER 639 38 Kärki. T. results are compared to other European countries, e.g. Finland, the marketing of wooden furniture differs greatly. In Finland, the furniture is marketed as good value for money, rich in selection and domestic (Pakarinen & Turunen 1999); whereas in Germany the design of the furniture, tree species and pro environmental aspects are highlighted. For solid-built furniture, the difference in marketing between Finland and Germany is even greater. When „green values" and tree species are used in the argumentation, the advertiser wants to highlight pro-environmental thinking. Altogether, in Central Europe considerable amounts of alder timber are used for furniture. In Study E, alder was defined as the fourth most popular tree species for all types of furniture. According to Euwid Magazin (Massivhölzer ... 1995), in 1995 alder was ranked as the sixth most popular tree species in German markets. This is consistent with the results of Study H, if we assume that, as seen from the import figures (Figures 4 and 5), the use of alder timber has increased. The prediction of branch-height characteristics, especially height to the first dead branch in Study 111, proved problematic, due to the poor correlation between these characteristics and other tree dimensions. This has also been found for other tree species (e.g. Rouvinen et ai. 1997, Verkasalo 1997 and Uusitalo and Kivinen 1998). When tree quality is considered, however, branch-height characteristics are important. Rouvinen et ai. (1997) proposed that average branch height characteristics should also be measured in field inventories when other stand characteristics are collected. These variables could then be generalized directly for all trees in the stand. This was also tested in Study 111 but the results were not good. Using non-parametric kernel-smoothing, Uusitalo and Kivinen (1998) developed a system for describing bivariate tree diameter and height to the distribution of the first dead branches. The benefits of this kind of approach are especially important in situations where the correlation structures between certain characteristics are complex (Silverman 1986). However, these non-parametric methods require measurement of several sample trees per stand and could thus not be used for the material of this study. When the reliability of the polynomial taper curve for grey alder (Model 3; V) was tested, the estimated relative bias and the standard error of the stem volume estimates were about 0.4 % and 9.0 %, respectively. Laasasenaho (1982) found that when the taper polynomial based on Fibonacci series was used, the estimated relative standard error of the volume estimates varied from 7.2 to 8.5 % for pine, spruce and birch. A good starting point for analysis of reliability is to compare volume estimates calculated with a taper polynomial to those estimated using a volume function. A simple non-linear volume function for grey alder has been presented in Study V using the same data as that used in Study HI for the taper polynomial. These two methods gave almost the same estimates of total stem volume. The taper polynomial was a more reliable and stable estimator, especially in the largest diameter classes. Rot is more common in grey alder than in spruce. Mäkelä (1990) studied the probability of rot in Norway spruce stems. If we compare the results of the Study ID to those of Mäkelä, some interesting differences can be seen. The Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland 39 proportion of rotten trees according to age classes was opposite in these two studies; in Norway spruce stands the proportion of rotten trees was smallest in the beginning of the rotation time, while in alder stands the proportion was smallest in the middle of the rotation time. The difference in alder stands could be explained as being due to the way alder regenerates (sprouting). There is another obvious difference in the diameter classes; the proportion of rotten Norway spruce is lowest in large diameter classes, while in alder stands the proportion of rotten trees is highest in large diameter classes. When results for the proportion of rotten trees at stump height in different age classes are compared to white birch (Betula pubescens) (Verkasalo 1997), the proportion of rotten grey alder stems in different age classes is significantly lower than in white birch. According to Verkasalo (1997), the proportion of rotten white birch trees (rot in the log or pulpwood) younger than 40 years ranged from 32 % (seed born) to 45 % (sprout born) (grey alder 36 %, Figure 2; III), at 41-60 years 37-53 % (grey alder 29 %) and over 61 years 45-61 % (grey alder 33 %). This is a significant result because it is normally thought that in Finland rot is much more common in grey alder than in other hardwood species. When Norokorpi (1979) investigated how the length of the rot in spruce stems correlates with Ö! 3 and the age of the tree, he found these correlations to be linear. In that study the volume of the trees and the volume of the rot were also compared and found to be exponential. Tamminen (1985) studied detection of butt-rot in standing spruces. The variables used described the form and growth of the stem of the sample trees (e.g. diameter increment at 1.3 m height during the last 10 years). The success rate in the prediction model was 68 %, which is comparable to the results of Study HI. Mäkelä (1990) also obtained similar results for spruce; the success rate for prediction of rot in a tree was 60 %. According to these studies, the success rates (73 and 67 %) found in Study ID, can be considered good. In general, according to Study IV, grey alder logs have knots from the base to the top. All types of knots appear and the length of the knot-free section at the base is short. The difference between a dry-knot section and a fresh-knot section is not as obvious as, for example, for pine or even birch (Kärkkäinen 1986). In small dimensioned logs, there are fewer knots than in larger logs. Especially in large top logs, there are many more fresh knots than in other types of logs. It is interesting that dry knots are concentrated at the base of the logs (0-0.5 m) and in the section from 1.5 to 2.5 m. The number of dry knots in top logs is almost twice that in the butt logs. Rotten knots are concentrated in the section from 0.5 to 2.5 m. Small-dimensioned butt and top logs are less knotty than large logs. If we compare these numbers of knots to the grade classification presented in Table 2 (IV), both averages could be included in grade-B. Large butt logs could be classified as grade-C and large top logs in the rejected category. Nevertheless, this is only a classification according to knottiness, in which the other quality factors (sweepness, crooks, cracks, discoloration etc.) are not taken into consideration. Overall, in different types of logs, it seems that the different grades are located in comparable sections along the length. It also seems that the worse the grade was, 40 Kärki, T. the longer was also the length of the grade. The most common reasons for decreasing grade were dry knots and discoloration. When the models for predicting the value of grey alder logs are considered, three models seemed to be valid: models predicting the value for butt logs (Model 1; IV), top logs including top diameter (Model 3; IV) and all logs including top diameter (Model 5; IV). Models 2 and 4 (IV) (based only on number of external knots) were also logical, but the standard errors were slightly higher because the top diameter was not included. The most important source of error in the material of Study IV was the subjective sampling of experimental trees by two loggers. Those trees obviously were not average trees in the stands as far as quality was concerned. On the other hand, according to the grades in Table 2 (IV), they were the best and largest trees. The measurements were made from only two stands. Along with the geographical concentration of the material, this permits only limited possibilities to apply the results elsewhere in Finland. Nevertheless, the material gives an indication of the quality of grey alder logs in the Savo-Karelia area. Unfortunately, the size of the knots (except classification) was not taken directly into account when the logs and timber were classified; if they had been, it would have permitted more detailed analysis of the knot distribution in different kinds of logs. The knots were measured only partly in order to classify the logs and sawn pieces correctly. The grading used for grey alder logs and sawn timber (Tables 2 and 3; IV) is also one source of error, because it is based on theoretical assumptions and the grade limits are fixed without empirical studies. This causes error when the grading is used for value predictions. Despite these possibilities for error, the reliability of the results can be considered good: one aim of the research was even to investigate the suitability of the classification (Tables 2 and 3; IV) to practise. In Article V, the theoretical calculated log volumes for grey alder were calculated according to three different classification criteria (Tables 6-9; V). It was concluded that the different dimension criteria greatly influence the theoretical estimates of log volume. When the volumes were estimated according to the minimum dimensions of Keinänen & Tahvanainen (1995) (Table 2; IV), the volumes were much smaller than when estimated with the criteria presented in Study I (Table 13; I); e.g. a grey alder stem of 14 cm diameter has to be 11 m tall to be able to produce a log with the dimension criteria 13 cm / 21 dm and even 18 m tall with dimension criteria 13 cm / 30 dm. With the dimension criteria 8 cm / 20 dm, a grey alder stem of 14 cm diameter produce a log as 7 m tall. It must be noted that the use of small dimensions in log criteria also means that the stem form must be straight so that the logs can be sawn. The bark volume of grey alder stems was also investigated in Study V. According to Table 10 and Model 10 (V), the proportion of bark increases as the diameter of a grey alder stem grows. The result is opposite to e.g. that of Mäkinen (1984). The difference could be explained so that in the study of Mäkinen the largest grey alders were less than 15 cm in diameter, whereas in Study V some stems were nearly 30 cm. Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland 41 The price from the calculation presented in Table 7, gave a stumpage price of 158 FIM/m3 . This result is a computational price based on average estimates. However, this price gives an indication of a reasonable price level for grey alder timber in Finland. It was also shown in the furniture manufacturing calculation that the raw material price for alder is not yet the limiting factor (Table 8). The price limit for good quality alder sawn timber could be 3 000 FIM/m3 . There are few studies and statistics concerning the prices of alder timber. In the study of Louna & Valkonen (1995) the computational price for alder timber for Finnish markets is calculated using the Swedish markets for hardwood logs in 1993 as a basis. The stumpage price for alder is calculated so that the price relation between birch and alder in Sweden was used for Finnish conditions using the Finnish birch price. The stumpage price of alder logs in Finland was calculated to be 152 FIM/m3 . In this case the stumpage price for birch in Finland was 200 FIM/m3 . If we assume that the price relations between hardwood logs in Sweden are nowadays the same as in year 1993, we can calculate the alder stumpage price by using the newer price for birch in Finland (263 FIM/m3). In this case, the computational stumpage price for alder in Finland would be 200 FEM/m 3, which agrees well with the calculated price in this study (Table 7). In year 1993, according to Louna & Valkonen (1995), the stock price for black alder was 150- 280 FIM/m3 and for grey alder 100-160 FIM/m 3 . In the same study, the price of unedged alder sawn timber was estimated to be 1 200 - 2 500 FIM/m3 . We can use the same methodology also for the German prices. It must be noticed that in this calculation we compare log prices from two different countries where the wood procurement and general price level is quite different. The price index between hardwood logs (B-quality) are calculated using the price levels in the state forests of Baden-Wiirttemberg and Bayern in 1997/98 (Jahresberichte der Baden-Wiirttembergische und Bayerische Staatsforsten 1998 and Metsätilastollinen vuosikirja 1998). The German prices for logs include logging, so to the Finnish price for birch sawlogs is added 50 FIM/m3 for logging costs (263 FIM/m3 +5O FIM/m3 = 313 FIM/m3). The birch prices are averages for silver and white birch, and the alder prices are average for black and grey alder. The calculated computational prices for hardwood logs in Finland in 1997/98 would then be for alder 417 FIM/m3 (index 133 compared to birch), beech 413 FIM/m3 (132), oak 635 FIM/m3 (203) and poplar and aspen 182 FIM/m 3 (58). This way calculated price for alder logs, 417 FIM/m 3 , seems to be very high compared to the price calculated in this research (Table 7). When the calculation for alder processing value is evaluated (Tables 5, 6, 7), attention must be focused on certain aspects. Firstly, the figures used in the calculations are average. Secondly, the processing value can vary significantly if the different values used in the calculation change. Thirdly, when alder is processed, by-products are very important; a decrease in pulpwood / firewood price significantly decreases the processing value of alder stems. Kärki, T. 42 7 CONCLUSIONS This dissertation covers many aspects of the procurement and mechanical processing of grey alder wood in Finland. These topics should, however, be studied more and developed further before a profitable level of utilization can be reached for grey alder. According to the results of this dissertation, there are potential markets for grey alder timber both in Finland and in Central Europe. The willingness of small- and middle-sized sawmills to process grey alder as one of their raw materials was a very significant finding. The results concerning the stand and wood quality of grey alder were also positive; there is a potential for mechanical wood processing of grey alder harvested from existing forests. This means that this species can be cultivated in Finland if grey alder forests are treated with goal-directed silvicultural methods, not selection thinned, as has been the practise until now. In addition, new information is given in this dissertation concerning the quality of grey alder stands and estimates of the saw log quality needed in practise to provide raw material for sawmills. The focus of this research was on grey alder forests and mechanical wood processing in eastern Finland, but the results can also be applied to certain conditions in other parts of the country. Moreover, the models for tree quality and volume can contribute indirectly to the increased utilization of alder timber. In conclusion, further research is needed if the use of grey alder is to be intensified. Relevant topics include: 1) properties and utilization of by-products, 2) organization and techniques for procurement of grey alder timber, e.g. in cooperation with a large wood supplier, 3) development of silvicultural methods for grey alder and 4) the mechanical and physical properties of grey alder for wood processing (e.g. drying, machining, gluing, surfacing). The management and utilization of grey alder as a profitable tree species is just beginning to develop in Finland. There is already an abundance of grey alder stands in certain parts of the country where stems could be raised to valuable timber. The use of alder must be seen as an addition to the main Finnish tree species, pine, spruce and birch. Grey alder timber could be one of the main materials used at small sawmills that concentrate on carefully selected special products. So, why not use silvicultural methods to raise this species and then process it for profit ? Grey aider (Alnus incana) as a raw material for mechanical wood processing in Finland 43 REFERENCES Alestalo, A. & Hentola, Y. 1967. Leppä sulfaattikeitossa. Paperi ja Puu nro 50: 25- 27. (In Finnish). Aravuo, K. 1994. Huonekalujen valmistus. Toimialaraportti 1994. KTM yrityspalvelu. Helsinki. 42 pp. (In Finnish). Außenhandel nach Waren und Ländern (Spezialhandel). Dezember und Jahr 1996. 1997. Statistisches Bundesamt, Wiesbaden. 452 pp. (In German). Björklund, T. & Ferm, A. 1982. Pienikokoisen koivun ja harmaalepän biomassa ja tekniset ominaisuudet. Folia Forestalia 500. 37 pp. (In Finnish). Bruun, H.H. & Slungaard, S. 1959. Investigation of porous wood as pulp raw material. Fibre dimensions of several NW European wood species. Paper and Timber No 2: 31-34. Farmer, R.H. 1981. Handbook of Hardwoods. Princes Risborough laboratory. London. 243 pp. Grönros, J., Merra, A. & Mali, J. 1995. Kotimaisten puulajien ominaisuudet ja saatavuus. Valtion teknillinen tutkimuskeskus. 59 pp. (In Finnish). Grosser, D. 1985. Die Hölzer Mitteleuropas. Springer-Verlag. Berlin. 208 pp. (In German). - 1989. Einheimische Nutzhölzer und ihre Verwendungsmoglichkeiten. Institut fiir Holzforschung der Universität Munchen. Miinchen. 46 pp. (In German). Hakkila, P. 1970. Basic density, bark percentage and dry matter content of grey alder (Alnus incana). Comm. Inst. For. Fenn. 71.5: 1-33. Handelsstrukturen fiir amerikanische Laubhölzer im Wandel. 1998. Holz- Zentralblatt 51-52: 788. (In German). Heino, M. 1988 a. Sahalaitosten kannattavuustarkastelu. Teknillinen korkeakoulu, puunjalostustekniikan laitos, puun mekaanisen teknologian laboratorio. Tiedonanto 47. 67 pp. (In Finnish). - 1988 b. Sahatavaratuotannon ja jatkojalostuksen kannattavuusvertailu. Teknil linen korkeakoulu, puunjalostustekniikan laitos, puun mekaanisen teknologian laboratorio. Tiedonanto 48. 20 pp. (In Finnish). Holmäsen, I. 1991. Pohjolan puut ja pensaat. WSOY. Helsinki. 177 pp. (In Finnish). Huonekalujen valmistus. 1996. KTM:n yrityspalvelun toimialaraportti. Kirjapaino Snellman Oy, Helsinki. 61 pp. (In Finnish). Huotari, P. 1995. Puuhuoltoverkoston organisointi ja kehittäminen. Joensuun yliopisto, metsätieteellinen tiedekunta. Tiedonantoja 28. 27 pp. (In Finnish). Huss-Danell, K. & Lundmark, J.-E. 1987. Growth of nitrogen-fixing Alnus incana and Lupinus spp. for restoration of degenerated forest soil in northern Sweden. Studia Forestalia Suecica 181. 20 pp. Hynynen, J. & Viherä-Aarnio, A. (ed.) 1999. Haapa - monimuotoisuutta metsään ja metsätalouteen. Asp - mängfald i skog och skogsbruk. Metsätutkimuslaitoksen tiedonantoja 725. 157 pp. (In Finnish). Ilvessalo, Y. & Ilvessalo, M. 1975. Suomen metsätyypit metsiköiden luontaisen Kärki, T. 44 kehitys-ja puuntuottokyvyn valossa. Acta Forestalia Fennica 144. 101 pp. (In Finnish). Jahresberichte der Baden-Wurttembergische Staatsforstverwaltung 1994-1998. Staatsministerium fiir Ernährung, Land-wirtschaft und Forsten. Stuttgart. (In German). Jahresberichte der Bayerische Staatsforstverwaltung 1991-1998. Staatsministerium fiir Ernährung, Land-wirtschaft und Forsten. Munchen. (In German). Juva, A. & Ahtorinne, M. 1995. Suomen huonekaluteollisuuden kilpailukyky ja keinot. Suomen ulkomaankaupan julkaisusarja 11/1995. 120 pp. (In Finnish). Kallio, M. & Salo, S. 1992. Puupörssi. Helsingin kauppakorkeakoulun julkaisuja D-156. Helsinki. 44 pp. (In Finnish). Kärki, T. 1995. Erikoispuutavaran käsitteen analyysi ja sen empiirinen sovellus Savo-Karjalassa. Metsäteknologian syventävien opintojen tutkielma. Joensuun yliopisto. 89 pp. (In Finnish). - 1997. Haapa- ja leppätukkien kysyntä, hankinta ja laatu. Joensuun yliopisto, metsätieteellinen tiedekunta. Tiedonantoja 53. 78 pp. (In Finnish). Kärkkäinen, M. 1986. Malli männyn, kuusen ja koivun puuaineen oksaisuudesta. Silva Fennica 20(2): 107-116. (In Finnish with English summary). Keinänen, E. & Tahvanainen, V. 1995. Pohjolan jalot puut. Pohjois-Savon erikoispuiden käytön lisäämisprojekti. 160 pp. (In Finnish). Kline, M. 1987. Alnus rubra - Red alder. In A Guide to Useful Wood of the World, Flynn Jr., J.H., Editor. King Philip Publishing Co., Portland, Maine, p. 34- 35. Kohonen, T. 1997. Pienten teollisuusyritysten vientikykyisyys - tutkimuskohteena kalusteala. Helsingin kauppakorkeakoulu, lisensiaatintutkimus. 184 pp. (In Finnish). Kräftiger Anstieg der deutschen Laubholzimporte. 1998. Holz-Zentralblatt 59: 902. (In German). Kuukka, J. 1980. Sahalaitoksen pienpuulinjan investoinnin kannattavuusanalyysi. Diplomityö, Lappeenrannan teknillinen korkeakoulu. 115 pp. (In Finnish). Kuula, M., Kallio, M., Salo, S. & Vepsäläinen, A. 1992. Puupörssi-simulaattori käyttöohje. Helsingin kauppakorkeakoulun julkaisuja D-158. Helsinki. 20 pp. (In Finnish). Laasasenaho, J. 1982. Taper curve and volume functions for pine, spruce and birch. Comm. Inst. For. Fenn. 95(8): 1-63. - & Snellman, C-G. 1983. Männyn, kuusen ja koivun tilavuustaulukot. Metsäntutkimuslaitoksen tiedonantoja 113. 91 pp. (In Finnish). Lahti, A. 1991. Pohjoismaisen huonekaluteollisuuden kilpailuasetelma. Helsingin kauppakorkeakoulu, työpapereita. F-289. 24 pp. (In Finnish). Laidlaw, W.B.R. 1960. Guide to British Hardwoods. Published by Leonard Hill Limited, 9 Eden Street, N.W.I, London. Lallukka, H. 1986. Puutavaran hinnan määrääminen. Tapion taskukirja, 20. painos, p. 428-434. (In Finnish). Lehtonen, 1., Pekkala, O. & Uuusvaara, O. 1978. Tervalepän (Alnus glutinosa (L.) Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland 45 Gaertn.) ja raidan (Salix caprea L.) puu- ja massateknisiä ominaisuuksia. Folia Forestalia 344:1-19. (In Finnish). Lönnberg, B. & Vesikallio, H. 1980. Sahapuuta pienempien puunosien suhteellinen arvomassa- ja levyteollisuudessa sekä lämpöenergian tuotannossa. Metsätehon tiedotus 366. 10 pp. (In Finnish). Louna, T. & Valkonen, S. 1995. Kotimaisen raaka-aineen asema lehtipuiden teollisessa käytössä. Metsäntutkimuslaitoksen tiedonantoja 553. 38 pp. (In Finnish). Lutz, J.F. 1978. Wood veneer: Log selection, cutting and drying. U.S. Forest Service, technical bulletin no 1577. 137 pp. Mäkelä, H. 1990. Tyvilahon esiintymistodennäköisyys etelärannikon kuusissa ja lahon korkeuden ennustaminen. Metsänarvioimistieteen pro gradu - tutkielma. Joensuun yliopisto. 51 pp. (In Finnish). Mäkelä, M. 1986. Lehtipuun korjuu kuormainharvesterilla Pohjois-Suomessa. Metsäteho. Helsinki. 6 pp. (In Finnish). Mäkinen, T. 1984. Harmaalepän ja tervalepän runkomuoto, kuorimallit ja runkokäyrämallit. Metsänarvioimistieteen pro gradu -tutkielma, Helsingin yliopisto. 60 pp. (In Finnish). Mali, J. 1980. Kotimaisten puulajien ja tuontipuulajien tekniset ominaisuudet ja käyttö. Valtion teknillinen tutkimuskeskus, puulaboratorio. Tiedonanto 3. Espoo. 43 pp. (In Finnish). Maltamo, M., Puumalainen, J. and Päivinen, R. 1995. Comparison of Beta and Weibull functions for modelling basal area diameter distribution in stands of Pinus sylvestris and Picea abies. Scandinavian Journal of Forest Research 10: 284-295. Massivhölzer: Buche an der Spitze - Lenga in Kommen. 1995. Euwid Magazin 4: 28-29. Metsätilastollinen vuosikirja 1998. SVT Maa- ja metsätalous 1998:3. 344 pp. (In Finnish). Miettinen, L. 1933. Tutkimuksia harmaalepiköiden kasvusta. Metsätieteellisen tutkimuslaitoksen julkaisuja 18.1:1-100. (In Finnish with German summary). Niemiec, S.S., Ahrens G.A, Willits S. and Hibbs D.E. 1995. Hardwoods of the Pacific Northwest. Oregon State University, College of Forestry, Research Contribution 8, Forest Research Laboratory, Department of Forest Products, Corvallis, Oregon. Norokorpi, Y. 1979. Old spruce stands, amount of decay and decay-causing microbes in northern Finland. Comm. Inst. For. Fenn. 97(6). 77 pp. Oijala, T. 1994. Puunkorjuun ja puutavaran kaukokuljetuksen tilastolukuja vuodelta 1993. Metsäteho. 6 pp. (In Finnish). Oijala, T., Vastamäki, A. ja Örn, J. 1994. Korjuun kustannusvertailun laskentaohjelman käyttö - Excel. Metsäteho. 8 pp. (In Finnish). Pakarinen, T. & Turunen, K. 1999. Puuhuonekalujen ja -huonekaluteollisuuden menestystekijät. Joensuun yliopisto, metsätieteellinen tiedekunta. Tiedonantoja 88. 68 pp. (In Finnish). Kärki, T. 46 Pöyhönen, I. 1991. Suomen sahateollisuuden kehityksen keskeiset muutokset vuoteen 2000. Teknillinen korkeakoulu, puunjalostustekniikan laitos, puun mekaanisen teknologian laboratorio. Tiedonanto 60. 216 pp. (In Finnish). Press, W.H., Teukolsky, S.A., Vetterling, W.T. and Flannery, B.P. 1992. Numerical Recipes in Fortran. The Art of Scientific Computing. Second edition. University Press, Cambridge. 963 pp. Routala, O. & Sihtola, H. 1934. Tutkimuksia lepän käyttömahdollisuuksista selluloosan raaka-aineena. Acta Chemica Fennica 7: 113-119. (In Finnish). Rouvinen, S., Kangas, A. and Maltamo, M. 1997. Männikön laatujakauman kuvaaminen oksarajatiedon avulla kuvioittaisessa arvioinnissa. Metsätieteen aikakauskirja -Folia Forestalia 4/1997:477-492. (In Finnish). Ryti, N. 1987. Puunjalostustalous. Otakustantamo. Espoo. 119 pp. (In Finnish). Saarsalmi, A., Palmgren, K. & Levula, T. 1985. Leppäviljelmän biomassan tuotos sekä ravinteiden ja veden käyttö. Folia Forestalia 628. 24 pp. (In Finnish). - , Palmgren, K. & Levula, T. 1991. Harmaalepän vesojen biomassan tuotos ja ravinteiden käyttö. Folia Forestalia 768. 25 pp. (In Finnish). - , Palmgren, K. & Levula, T. 1992. Harmaalepän ja rauduskoivun biomassan tuotos ja ravinteiden käyttö energiapuu viljelmällä. Folia Forestalia 797. 29 pp. (In Finnish). Salmi, J. 1977. Suomalaisia ja ulkolaisia puulajeja. Osa II: Lehtipuut A-N. Helsingin yliopisto, metsäteknologian laitos. Tiedonantoja 35. Helsinki. 282 pp. (In Finnish). - 1978 a. Suomalaisia ja ulkolaisia puulajeja. Osa HI: Lehtipuut O-Ö. Helsingin yliopisto, metsäteknologian laitos. Tiedonantoja 38. Helsinki. 298 pp. (In Finnish). - 1978 b. Tervalepän puuaine ia sen käyttö. Dendrologian seuran tiedotuksia 9(3): 75-78. (In Finnish). Silverman, B.W. 1986. Density estimation for statistics and data analysis. Chapman & Hall. 175 pp. Siikaluoma, S.O. 1991. Terminaalikauppa: hankintakauppojen kehittä misvaihtoehto ja sen liiketaloudellinen kannattavuus. Helsingin yliopisto, metsätalouden liiketieteen laitos. Pro gradu -työ. 61 pp. (In Finnish). Siitonen, M. 1990. Pohjois-Karjalan metsien kehitysvaihtoehdot. Metsäntutkimuslaitoksen tiedonantoja 357: 45-63. (In Finnish). Tamminen, P. 1985. Butt rot in Norway spruce in southern Finland. Comm. Inst. For. Fenn. 127. 52 pp. Tikka, P.S. 1954. Haapametsiköiden rakenteesta ja laadusta. I. Rakenne. Comm. Inst. For. Fenn. 43: 1-33. (In Finnish). USDA. 1987. Wood Handbook: Wood as an Engineering Material. Agriculture Handbook No. 72. United States Department of Agriculture, Forest Service, Madison, Wisconsin. Uusitalo, J. & Kivinen, V.-P. 1998. Constructing bivariate DBH/dead branch height distribution of pines for use in sawing production planning. Scandinavian Journal of Forest Research 13(4): 510-515. Valkonen, S.. Rantala, S. & Sipilä A. 1995. Jalojen lehtipuiden ja tervalepän Grey alder (Alnus incana) as a raw material for mechanical wood processing in Finland 47 viljelyjä kasvattaminen. Metsäntutkimuslaitoksen tiedonantoja 575. 112 pp. (In Finnish). Vepsäläinen, A. & Kuula, M. 1992. Puupörssin toteuttaminen ja vaikutukset puukaupan logistisiin palveluihin. Helsingin kauppakorkeakoulun julkaisuja D-157. Helsinki. 29 pp. (In Finnish). Verkasalo, E. 1993. Lehtikuusen mittaus ja kauppa. Metsäntutkimuslaitoksen tiedonantoja 464: 91-98. (In Finnish). - 1997. Hieskoivun laatu vaneripuuna. Metsäntutkimuslaitoksen tiedonantoja 632. 483 pp. (In Finnish). Vuokila, Y. 1977. On the growth capacity of aspen stands on good sites. Folia Forestalia 299. 11 pp. (In Finnish with English summary) Wagenfiihr, G. 1996. Holz-Atlas. 4. neuarbeitete Auflage. Fachbuchverlag Leipzig. 688 pp. tutkimusartikkeliJ \ ••{ Afcofl Af AP n ,v 1 V IOCI LI vlvv J t aikakauskirja 37 Timo Kärki Sahauskelpoisen erikoispuun laatuvaatimukset ja käyttö Savo- Karjalan alueella Kärki, T. 1997. Sahauskelpoisen erikoispuun laatuvaatimukset ja käyttö Savo-Karjalan alueella. Metsätieteen aikauskirja - Folia Forestalia I /1997: 37-48. Tutkimuksessa on analysoitu Suomessa vähemmän käytettyjen puulajien, ns. erikoispuu lajien, laatu- ja mittavaatimuksia sekä käyttöä Savo-Karjalan alueella. Työssä käsitellyt puu lajit ovat laadukas mänty, kuusi ja koivu sekä lehtikuusi, kataja, haapa, leppä, pihlaja ja Suomessa luontaisena kasvavat jalot lehtipuut Erikoispuulajeille muodostettiin Savo-Karjalan alueelle tehdyn yrityskyselyn perusteella kolmiluokkainen laatu- ja mittaluokittelu (luokat A, B ja C). Laatuluokittelussa on käytetty sanallista arviointia vikaisuuksien määrittelyssä ja mittavaatimuksissa on annettu kussakin luokassa minimiläpimitta- ja minimipituusvaatimukset tavaralajimenetelmällä valmistetulle erikoispuuta varalle. Yrityskyselyn tulosten perusteella suhtautuminen erikoispuun käyttöön Savo-Karjalassa on positiivisen kiinnostunutta. 70 % kyselyyn vastanneista yrittäjistä on kiinnostunut laajen tamaan tuotantoaan erikoispuulajeihin, jos niitä on vain saatavissa. Kiinnostavimmat puulajit ovat leppä sekä haapa. Pääkomponenttianalyysilla saatiin neljä erilaista tekijää, jotka ovat ratkaisevia erikoispuun käytön aloituksessa tai käytön lisäämisessä (uusi tuotantomahdolli suus, erikoispuunhankinta, markkinat, kehittämisprojektit). Vastanneet yrittäjät haluavat kehittää puunhankintaansa tulevaisuudessa omana hankintanaan tai metsänomistajan suo raan toimittamana. Asiasanat: erikoispuulaji, puupörssi, laatuvaatimukset Kirjoittajan yhteystiedot; Timo Kärki, Joensuun yliopisto, metsätieteellinen tiedekunta, PL 111, 80101 Joensuu. Faksi (013) 251 4444, sähköposti Timo.Karki@forest.joensuu.fi Hyväksytty 1 1.2.1997 Tutkimusartikkeli Metsätieteen aikakauskirja 1/1997 38 1 Johdanto 1.l Yleistä Suomessa puun teollinen käyttö on keskittynyt kol meen suureen pääpuulajiin eli mäntyyn, kuuseen ja koivuun. Muiden kotimaisten puulajien käyttö on ollut huomattavasti vähäisempää. Haapaa ja leppää on käytetty aikakaudesta riippuen vaihtelevia mää riä, mutta muiden puulajiemme kuten pihlajan, tam men, saarnen, jalavan, vaahteran ja lehmuksen käyt tö on ollut hyvin vähäistä (Louna ja Valkonen 1995). Vähemmän käytettyjen puulajien kohdalla onkin puhuttu erikoispuulajeista ja erikoispuista. Markkinoilla liikkuvan kotimaisen erikoispuun määrä on marginaalinen. Eniten käytetyt erikois puulajit ovat visakoivu, haapa, leppä sekä lehtikuu si. Jalojen lehtipuiden kohdalla kotimaisen raaka aineen määrä on rajoitettu, yhteismäärältään alle 200 m 3 pyöreää puuta vuodessa (taulukko 1). Sa manaikaisesti lehtipuusahatavaraa (poislukien koi vu) tuotiin Suomeen (v. 93) noin 35 000 m 3 (Louna ja Valkonen 1995). Annettuihin lukuihin on kuitenkin syytä suhtau tua varauksella, koska Lounan ja Valkosen (1995) tutkimuksessa olivat mukana vain yli 20 työnteki jän yritykset sekä pienemmistä yrityksistä satun naisesti 10 % otos. Viime vuosina on mielenkiinto lisääntynyt harvi naisempien kotimaisien puulajien käyttämiseen ra kentamisessa ja kodin sisustuksessa. Aiheesta on käyty vilkasta keskustelua alan lehdissä ja useita erikoispuun käytön kehittämistä edistäviä hankkei ta on käynnistetty (mm. Puu-Suomi -projektissa 1992-1994). Alueellisiin kehittämisprojekteihin on liittynyt myös erikoispuupörssin perustaminen, jon ka avulla on tehty pienimuotoista erikoispuun kaup paa. Tällaisia puupörssejä on toiminnassa lisalmen seudulla, Pohjois-Kaijalassa, Itä-Hämeessä, Sata kunnassa ja Länsi-Uudellamaalla. Kuitenkin esim. Pohjois-Karjalan Erikoispuupörssin toimintaa on vaikeuttanut erikoispuun yleisten laatu- ja mitta vaatimusten puuttuminen (Saramäki Matti, Poh jois-Karjalan mhy:sten liitto 1994). Puupörssin to teuttamista ja vaikutuksia puukaupan logistisiin pal veluihin ovat tutkineet Vepsäläinen ja Kuula (1992) julkaisussaan ja puupörssin toimintaan ovat paneu Taulukko 1. Erikoispuun käyttömäärät vuonna 1993. Mukana ei ole polttopuukäyttöä. (Verkasalo 1993 a, Louna ja Valkonen 1995) tuneet Kallio ja Salo (1992) ja Kuula ym. (1992). Erikoispuu-käsitteeseen liittyvää suomalaista tut kimuskirjallisuutta on ollut saatavilla niukasti. Kat tavia puulajien ominaisuuksia käsitteleviä perus teoksia on julkaistu (Salmi 1972, 1977, 1978, Uus vaara ja Pekkala 1979, Mali 1980), mutta itse eri koispuun käytöstä puuttuvat kattavat tutkimukset. Lehtikuusen kohdalla tilanne on poikkeava, sillä lehtikuusesta on tehty Suomessa muita erikoispuu lajeja huomattavasti kattavampi tutkimusten sarja jo melko pitkän ajan kuluessa (Vuokila 1960, Hak kila ja Winter 1973, Sairanen 1982, Tuimala 1993, Verkasalo 1993 a, 1993 b). Haavan ominaisuuksia ovat tutkineet Tikka (1954, 1956), Kärkkäinen (1980, 1981), Kärkkäinen ja Salmi (1978), Opdahl (1989, 1991), Verkasalo (1990), leppää Schalin (1966), Alestalo ja Hentola (1967), Lehtonen ym. (1978), Björklund ja Ferm (1982) ja visakoivua Saarnio (1976). Aivan viime aikoina on kiinnostus muidenkin erikoispuulajien käyttöön herännyt ja muutamia julkaisuja aiheesta on tehty (Palen 1994, Keinänen ja Tahvanainen 1995, Louna ja Valko nen 1995, Valkonen ym. 1995). Suomessa erikoispuun käyttöä on tarkasteltu muu tamassa tutkimuksessa. Louna ja Valkonen (1995) toteavat lehtipuiden käytöstä, että teollisuussahat asettavat lehtipuutukeille yleensä samankaltaisia laatuvaatimuksia kuin havupuutukeille. Pienyrittä jät sen sijaan kelpuuttavat sahattaviksi myös huo Puulaji Käyttömäärä Visakoivu Muutamia satoja tuhansia kiloja Haapa Haapatukki: 10 000-15 000 m 3 Haapakuitu: 100 000-150 000 m 3 Leppä Leppätukki: Muutamia tuhansia m 3 Leppäkuitu: Muutamia kymmeniä tuhansia m 3 Lehtikuusi 1 000-2 000 m 3 Tammi Alle 100 m 3 Pihlaja Muutamia kymmeniä m 3 Vaahtera Muutamia kymmeniä m 3 Lehmus Muutamia kymmeniä m 3 Saarni Muutamia m 3 Jalava Muutamia m 3 Kärki Sahauskelpoisen erikoispuun laatuvaatimukset... 39 Taulukko 2. Luttisen (1994) tekemässä tutkimuksessa annettuja mittavaatimuksia haavalle ja lepälle. Osassa järeys- ja pituusmitoista on annettu vain minimimitat eikä maksimidimensiolle ole annettu ylärajaa. nolaatuisia ja pienikokoisia tukkeja. Minimimitoiksi lehtipuutukeille annetaan tutkimuksessa yleisesti läpimitan osalta 14-20 cm sekä pituuden osalta 1,5-3 m. Luttinen (1994) on tutkinut Ylä-Savon puupörs siä varten kyselytutkimuksella yrittäjien mielestä eri puutavaralajeille soveliaimpia järeysasteita ja pituuksia. Tutkimuksessa olivat mukana myös lep pä ja haapa. Pienimuotoisessa tutkimuksessa (vas taajia 21) muutamien yrittäjien antamia mittavaati muksia esitellään seuraavassa taulukossa (taulukko 2). Tutkimuksen mukaan yrittäjät haluavat haapa tukin järeämpänä kuin leppätukin. Tämä johtunee osittain siitä, että harmaaleppä jää jo luonnostaan pienemmäksi kuin haapa ja tervaleppä. Pituuden osalta yrittäjät ovat melko yksimielisiä 3,0 m mini mipituudesta. Erikoispuiden laatu-ja mittavaatimuksia käsittele viä tutkimuksia löytyy lähinnä saksankielisestä kirjallisuudesta. Laatutunnuksia ovat puulajeittain ja käyttötarkoituksittain tarkastelleet Knigge ja Schulz (1966), Steuer (1979), Grammel (1989) ja pituus-ja läpimittavaatimuksia Eisele (1989 a, 1989 b, Roh holzsortierungsbestimmungen ... (1993). Saksalai sesta luokituksesta on huomattava, että luokitus kes kittyy runkomenetelmällä apteerattujen runkojen ar viointiin. 1.2 Erikoispuulajin ja -puutavaralajin käsite Metsissämme esiintyy luontaisena kolme havupuu ja neljätoista lehtipuupuulajia: euroopanmaijakuu si, mänty, kuusi, haapa, rauduskoivu, hieskoivu, tervaleppä, harmaaleppä, metsätammi, vuorijala va, kynäjalava, kotipihlaja, suomenpihlaja, ruot sinpihlaja, vaahtera, metsälehmus ja saarni (Salmi 1972, Holmäsen 1991). Osa näistä puulajeista on hyvin rajoitetusti tietyllä kasvualueella esiintyviä, joka sinällään rajaa ko. puulajit erikoispuiksi. Suomeen on aikojen kuluessa tuotu vierasta alku perää olevia puulajeja, joita esiintyy nykyisin maas samme. Näistä puulajeista taloudellisesti merkittä vimpiä ovat euroopan- ja siperianlehtikuusi. Näi den puulajien alkuperä ei ole kotimainen, joten ne täyttävät toisen erikoispuu -käsitteen rajaavan teki jän eli esiintyvät luontaisen kasvualueensa ulko puolella ja ovat sopeutuneet fysiologisesti maas samme vallitsevaan ilmastoon. Lisärajoite erikoispuu -käsitteelle on ko. puulajin pieni osuus metsäpinta-alasta ja vuotuisesta kas vusta. Männyn osuus Suomen metsäpinta-alasta vallitsevana puulajina on 61,9 % ja kuusen 27,1 % (Metsätilastollinen vuosikirja 1992). Metsiköitä, joissa koivu on vallitsevana puulajina on 7 %. Le pän ja haavan osuudet ovat 0,5 % ja 0,3 %. Muille havu- ja lehtipuille on metsälautakunnittain saatu prosentin kymmenyksen suuruisia arvoja; tilavuu della mitattuna erikoispuulajien suuruusjärjestys hieman muuttuu. Louna ja Valkonen (1995) ovat laskeneet Valtakunnan metsien 8. inventoinnin (VMI8) perusteella Etelä-Suomen lehtipuun koko naisvarannot ja tukkivarannot puulajeittain (tau lukko 3). Taulukosta nähdään, että haapa ja leppä ovat ko konaistilavuudeltaan samaa luokkaa. Suhde leppä lajien sisällä on suuresti harmaalepän puolella. Koi vulajeissa hieskoivua on koivuvaroistamme mer kittävästi enemmän kuin rauduskoivua. Muiden lehtipuiden osuus kokonaispuuvaroistamme on hy vin pieni, alle 4 milj. m 3 Etelä-Suomen alueella. Vastaavasti esim. männyn kokonaistilavuus samal la alueella on 500,2 milj. m 3 (Metsätilastollinen vuosikiija 1992). Erikoispuu -käsitettä rajaava tekijä on myös puu aineen erikoiset rakenteet ja ominaisuudet. Tällai sia ominaisuuksia ovat visaisuus ja laineellisuus Puutavaralaji Järeysaste (cm) Pituus (m) Min Max Min Max Haapatukki 15 30 3,0 20 40 3,0 23 40 1,7 Leppätukki 8 - 3,0 6,1 13 40 2,7 Haapakuitu 15 30 3,0 Leppäkuitu 8 - 2,0 3,0 15 30 3,0 Tirtkimusartikkeli Metsätieteen aikakauskirja 1/1997 40 Taulukko 3. Etelä-Suomen lehtipuuvarannot Valtakun nan metsien 8. inventoinnin (VMI8) mukaan. Määrät 1000 m 3. (Etelä-Suomi käsittää Ahvenanmaan ja 15 ete läisintä metsälautakuntaa, joihin kuuluvat myös Pohjois- Karjala ja Pohjois-Savo.) eri puulajien puuaineksessa sekä puunrungoilla kas vavat pahkat, jotka voidaan laskea puuntuotannon sivutuotteiksi. Visaisuutta esiintyy, koivun visai suuden lisäksi, myös terva- ja harmaalepällä, tam mella, männyllä sekä kuusella (Saarnijoki 1961). On huomattava, ettei kuusen ja männyn visa ole sama asia kuin mukuramuodostumat ko. puitten rungoissa, ns. mukuramännyt (Ahola 1952, Saarni joki 1961). Soitinten valmistukseen soveltuvat puut voidaan laskea myös erikoispuiksi. Hyviä soitin puita ovat vanhat hitaasti kasvaneet jättiläiskuuset, pihlajat, tervalepät, haavat, vaahterat ja koivut, joilla on hyvät soitinpuulle ominaiset ominaisuudet, mm. kumajavuus. Erikoispuulajit ovat siis edellä mainitun perus teella puulajeja, jotka esiintyvät rajoitetusti luon taisen kasvualueensa ulkopuolella, omaavat pienen osuuden metsäpinta-alasta ja vuotuisesta kasvusta sekä ovat rakenteeltaan ja ominaisuuksiltaan poik keavia. Erikoispuulajien lisäksi voidaan puhua eri koispuutavaralajeista, jotka voivat olla mitä puula jia tahansa, mutta lasketaan käyttötarkoituksen, eri koisten dimensioiden tai vähäisen käyttömäärän vuoksi erikoispuutavaralajiksi. Erikoispuutavaraa on käyttökohteen edellyttämiin mittoihin sahattu ja katkottu puutavara, kuten kaivospuut, laatikkolauta puut, rullapuut, pylväät, junttapaalut, ratapölkyt sekä egyptin parrut (Lakio 1953). Ko. puutavaralaji ei myöskään mahdollisesti sovellu nykyiseen puu tavaraluokitteluun. 1.3 Tutkimuksen tavoitteet Tässä tutkimuksessa selvitetään Suomessa vähem män käytettyjen puulajien laatu- ja mittavaatimuk sia ja niiden nykyistä ja potentiaalista käyttöä Savo- Kaijalan alueella. Tutkimustyössä perehdytään seuraavien erikois puulajien käyttöön: - laadukas mänty (Pinus sylvestris) - laadukas kuusi (Picea abies) - laadukas koivu ( Betula pendula ja B. pubescens ); visakoivu - lehtikuusi (Larix decidua ja L. sibirica) - leppä (Alnus incana ja A. glutinosa) - haapa (Populus tremula) - kataja (Juniperus communis) - pihlaja (Sorbus aucupariä) - saarni (Fraxinus excelsior) - tammi (Quercus robur) - vaahtera (Acer platanoides) - lehmus (Tiliä cordata) - jalava ( Ulmus glabra ja U. laevis) Tutkimuksen tavoitteet jakautuvat kahteen eri osaan: 1. Muodostetaan erikoispuutavaralajeille laatu-ja mitta vaatimukset kyselyn perusteella. 2. Kartoitetaan erikoispuun nykyistä ja potentiaalista käyttöä Savo-Kaijalan alueella. Tutkimus on tehty Joensuun yliopiston sekä Metsä miesten Säätiön myöntämien apurahojen turvin. Professorit Pertti Harstela sekä Matti Kärkkäinen, lehtori Lauri Sikanen ja MH Pekka Huotari lukivat käsikirjoituksen ja tekivät siihen arvokkaita paran nusehdotuksia. Esitän parhaat kiitokseni kaikille tutkimuksen valmistumiseen myötävaikuttaneille henkilöille ja organisaatioille. 2 Aineisto ja menetelmät Tutkimuksen aineistona käytettiin helmikuussa 1994 kerättyä postikyselyaineistoa. Tutkittavien yritysten yritystiedot (toimiala, postiosoite) saatiin Puulaji Tukkipuu Kuitupuu Yhteensä Rauduskoivu 15 828 36 877 52 705 Hieskoivu 11 193 96 731 107 924 Haapa 2 348 13 266 15 614 Tervaleppä 245 3 362 3 607 Harmaaleppä 69 11 457 11 526 Muu lehtipuu 67 3 713 3 780 Yhteensä 29 750 165 406 195 156 Kärki Sahauskelpoisen erikoispuun laatuvaatimukset... 41 Kuva 1. Yrityskyselyyn vastanneiden puuteollisuusyritysten toimialajakauma Savo-Karjaiassa. maakuntaliittojen ylläpitämästä teollisuusyritys rekisteristä. Tutkimuskohteena olivat kattavasti kaikki Pohjois-Karjalan ja Pohjois-Savon pyöreää puuta sahaavat sahayritykset (201 kpl). Lisäksi ky sely lähetettiin osalle alueen puusepänyrityksistä (136 kpl). Nämä yritykset valittiin sillä perusteella, että ne käyttävät raaka-aineenaan pyöreää puuta, mutta eivät esimerkiksi levytuotteita. Myös alueel la sijaitseville vaneritehtaille (2 kpl) lähetettiin ky selylomake, joilta todettiin, etteivät ko. tehtaat käytä tutkimuksen piiriin luettavia erikoipuutavaroita tuo tannossaan. Tehtyyn kyselyyn vastasi määräaikaan mennessä 52 yritystä. Kahden viikon kuluttua kyselyn lähet tämisestä postitettiin karhukirje kaikille niille yrit täjille, jotka eivät olleet vielä vastanneet kyselyyn. Tämän jälkeen kyselyyn vastanneita yrityksiä oli 84 kappaletta, jolloin vastausprosentiksi muodos tuu 25 %. Näistä yrityksistä 40 sijaitsee Pohjois- Karjalan läänin alueella ja 44 yritystä Kuopion lää nin alueella. Vastanneiden puuteoHisuusyritysten toimialajakauma koko Savo-Karjalan alueella ja kautuu kuvan 1 mukaisesti. Menetelminä tutkimuksessa on käytetty taulu kointia sekä pääkomponenttianalyysia. Taulukoin tia on käytetty yksinkertaisten yhteenvetotaulukoi den luontiin, joista ilmenee nopeasti määrällisiä eroavaisuuksia. Monimuuttujamenetelmistä pää komponentti analyysia käytettiin aineiston analy sointiin. Monimuuttujamenetelmään päädyttiin, koska haluttiin pelkistää monimutkaisen ja laajan aineiston tietoa. Pääkomponenttien koostumusta selvitettäessä oli vat periaatteet seuraavat: kertoimen suuri lukuarvo kertoo kyseisen muuttujan vaikuttaneen voimak kaasti pääkomponentin muodostukseen ja etumerkki kertoo vaikutuksen suunnan (Ranta ym. 1992). Ana lyysista saaduille komponenteille tehtiin vielä nk. Varimax-rotatointi, eli pyrittiin helpottamaan kom ponenttien tulkintaa niin, että muuttujien lataukset olivat joko selvästi pieniä tai suuria (Biihl 1994). Analyysin tuloksena saatujen pääkomponenttien koostumus selvitettiin ja annettiin komponenteille sanalliset kuvaukset. 3 Tulokset 3.1 Laatu-ja mittavaatimukset Yrityksiä pyydettiin asettamaan laatu-ja mittavaa timuksia heitä kiinnostaville puulajeille. Laaduk kaalle männylle, kuuselle ja koivulle asetetut mitta vaatimukset eivät poikenneet merkittävästi jo käy Tutkimusartikkeli Metsätieteen aikakauskirja I /1997 42 Taulukko 4. Vastanneiden yrittäjien asettamia laatu vaatimuksia männylle, kuuselle ja koivulle Taulukko 5. Vikaisuuksia puutavaralajeissa sallivien yri tysten suhteellinen osuus. tössä olevista. Sen sijaan käyttäjät asettivat taulu kon 4 mukaisia laatuvaatimuksia ko. puulajeille. Männyn kohdalla vähäoksainen tukki on halu tuinta sahattavaa, kun taas kuuselle on menekkiä normaalilaatuisena sahatukkina. Koivun osalta yri tykset ovat halukkaimpia ottamaan vastaan oksa tonta tyvitukkia. Suhtautuminen lenkouteen, lahoon ja oksakyhmyihin on taulukon 5 mukaista. Lievä lenkous sallittiin kahdessa kolmannekses sa yrityksistä joka puulajin kohdalla. Lahoa ei sal littu männyssä eikä kuusessa. Oksakyhmyt raaka aineessa hyväksyy neljännes männyn ja kuusen ja lostajista. Koivulla hyväksymisprosentti on suuri sekä lahon että oksakyhmyjen kohdalla. Lehtikuuselle, lepälle ja haavalle asetettiin taulu kon 6 mukaisia laatu- ja mittavaatimuksia. Lehtikuuselle yrittäjät asettivat normaalin saha tukin laatuvaatimukset. Läpimittavaatimukset ovat hieman alhaisemmat kuin männyllä ja kuusella. Haapaa ja leppää yrittäjät hyväksyvät myös hei kompilaatuisena tukkina, jonka vähimmäisläpi mitaksi on asetettu haavalla 14 cm ja lepällä 13 cm. Taulukko 6. Yrittäjien asettamia laatu- ja mittavaati muksia lehtikuuselle, haavalle ja lepälle. Taulukko 7. Vastanneiden yrittäjien asettamia pituus vaatimuksia haavalle ja lepälle. Puutavaralajin pituudella ei yrittäjien mielestä ole suurta merkitystä lehtikuusen kohdalla. Sen si jaan haavalle ja lepälle asetettiin taulukon 7 mukai sia kriteerejä. Haavalle ja lepälle asetetut pituusvaatimukset ovat samankaltaisia. Maksimipituudessa yrittäjät tahtoi vat kautta linjan pitempää leppää kuin haapaa. Muu toin kriteerinä pidettiin yleisesti sahauskelpoisuut ta. Lenkoutta sallittiin kaikissa kolmessa puulajis sa, mutta lahoa ei sallittu missään muodossa muu tamaa yrittäjää lukuun ottamatta. Oksakyhmyjä ei pidetty toivottavina lehtikuusessa; lepän ja haavan kohdalla oksakyhmyjä sallineiden osuus oli puolet vastanneista. Visakoivulle ja katajalle ei asetettu vaatimuksia läpimitan eikä pituuden suhteen, vaan haluttiin raa ka-ainetta rankoina sekä haarapölkkyinä. Pihlajaa haluttiin hitaasti kasvaneina tyvipölkkyinä (pituus 50-200 cm). 3.2 Erikoispuun käyttö Taulukosta 8 ilmenee, että harvinaisempien puu lajien käyttö on melko vähäistä määrällisesti, mut Laatuvaatimus Mänty Osuus vastaajista % Kuusi Koivu Oksaton tyvitukki 19 5 80 Vähäoksainen tukki 46 - - Normaalilaat. sahatukki 19 59 20 Välitukki 10 21 - Heikompilaatuinen, ei tukkipuun mittoja 6 15 - täyttävä puuaines Yhteensä 100 100 100 Vika Vikaisuuden sallineiden yritysten %-osuus Mänty Kuusi Koivu Lievä lenkous 70 79 60 Laho 4 5 40 Oksakyhmyt 21 26 40 Puulaji Järeysaste (cm) Laatuvaatimus Min Max Lehtikuusi 14 55 Normaalilaatuinen sahatukki Haapa 14 Heikompilaatuinen sahatukki Leppä 13 Heikompilaatuinen sahatukki Puulaji Pituusvaatimus (m) Min Max Haapa 2,3 4,1 Leppä 2,5 5,2 Kärki Sahauskelpoisen erikoispuun laatuvaatimukset... 43 Taulukko 8. Harvinaisempien puulajien käyttö Savo- Karjalassa sekä arvio tarpeesta tulevaisuudessa. Taulu kon käyttömäärät on saatu 23 yrityksen vastausten sum mana. Mukana ei ole polttopuukäyttöä. Kuva 1. Eri puuteollisuusyritystyyppien kiinnostus eri koispuulajien käyttöön. Kiinnostus: ■ kyllä, □ ei. ta ainakin kyselyyn vastanneiden yrittäjien osalta harvinaisempia puulajeja käyttää yllättävän moni yrittäjä eli melkein joka kolmas. Erikoispuulajeja, etenkin haapaa, leppää ja lehtikuusta, haluttaisiin käyttää huomattavasti enemmän, jos niitä vain olisi saatavissa. Kotimaista pihlajaa käytetään alueen pienyrityksissä 7,5 m 3, kun Lounan ja Valkosen (1995) tutkimuksessa koko Suomen käytöksi on arvioitu muutamia kymmeniä kuutiometrejä. Suurin kiinnostus erikoispuulajien käyttöön on höyläystä suorittavilla sekä käyttö- ja koriste-esi neitä ja puusepänteollisuuden tuotteita valmistavil la yrityksillä (kuva 2). Osa näistä yrityksistä käyt tää jo tällä hetkellä huomattavia määriä erikoispuu ta tuotannossaan ja usea yritys olisi valmis lisää Taulukko 9. Rotatoitu pääkomponenttimalli yritys kyselyn muuttujista. mään käyttöä, jos raaka-ainetta olisi paremmin saa tavilla. Sahausta harjoittavista yrityksistä kaksi kol mesta on kiinnostunut erikoispuun tarjoamista mah dollisuuksista. Pääkomponenttianalyysilla analysoitiin, mitkä tekijät olivat yrittäjän mielestä ratkaisevia tekijöi tä, kynnyskysymyksiä, lisättäessä tai aloitettaessa erikoispuun käyttöä. Analysoinnissa käytettiin muuttujina kolmeatoista erilaista yritystä, sen puun käyttöä ja erikoispuunkäytön mahdollisuuksia ku vaavaa tekijää. Mallilla kyetään selittämään 56 % aineiston kokonaisvarianssista (taulukko 9). Tuloksena pääkomponenttianalyysista saatiin yksi koko tutkimuspopulaatiota selittävä komponentti (Kl), jonka selittävyys on suuri, 20 prosenttia. Neljä Puulaji Käyttö tällä hetkellä (m3/v) Arvioitu tarve (m 3 /v) Visakoivu 2 Muutamia kuutiometrejä Haapa 199 5 500 Leppä 297 5 000 Lehtikuusi 108 3 180 Kataja 4 Muutamia kuutiometrejä Pihlaja 7,5 Muutamia kymmeniä kuutiometrejä Jalot lehtipuut 15 Muutamia kymmeniä kuutiometrejä Yhteensä (m3/v) 632,5 Alle 14 000 Käyttömäärä / yritys (m 3 /v) 27,5 600 Muuttuja K1 K2 K3 K4 K5 Aloitus * 0,71 * * * Hankin 1 -* _* 0,46 _* -0,49 Kaytluo 0,84 _* * * * Kiinnost * 0,62 * * * Kuivaus -0,42 _* * -0,63 * Luokl 0,92 _* * _* _* Maara * * 0,30 0,36 * Markkina * _* _* 0,66 * Paapl -* 0,40 _* 0,37 _* Päätuote -* _* * 0,54 -0,38 Suunnitt -* 0,68 0,32 * -0,29 Tarpeell * * * * 0,45 Tietone * _* 0,51 _* * Ominaisarvot 2,38 1,99 1,54 1,36 1,18 Rotatoitujen komponenttien selitysaste (%) 20,0 10,3 9,4 8,6 7,4 aloitus = halukkuus aloittaa erikoispuutuotanto hankini = raaka-aineen hankintatapa nyt kaytluok = puun kokonaiskäyttö (m 3 /v) kiinnost = kiinnostus laajentaa erikoispuulajeihin kuivaus = tuotteen kuivaus luokl = henkilökunnan määrä maara = arvio tarvittavasta erikoispuulajin määrästä (m 3 /v) markkina = markkinoita erikoispuutuotteille paapl = yrityksen käyttämä pääpuulaji päätuote = yrityksen päätuote suunnitt = muutoksia suunnitteilla tuotantoon tarpeell = erikoispuulajien käytön lisäykseen ja tuotekehitykseen tähtäävien projektien tarpeellisuus tietone = tietoinen Enkoispuupörssistä Tutkimusartikkeli Metsätieteen aikakauskirja I /1997 44 muuta komponenttia selittävät tietyntyyppistä yrit täjää ja tekijää, joka on ratkaiseva erikoispuun käy tön aloituksessa tai käytön lisäämisessä. Kompo nentit ovat seuraavat: K 2: (Uusi tuotantomahdollisuus) Yrittäjä, joka on olosuhteiden vuoksi pakotettu suunnittelemaan muutoksia yrityksessään. Tämä lisää kiinnostusta erikoispuulajien käyttöön uutena tuotantomahdol lisuutena. K 3: (Erikoispuun hankinta) Yrittäjällä on jo tie toa erikoispuun käytöstä ja yritys saattaa käyttää erikoispuulajeja jalostuksessaan. Erikoi spuunhan kinta nousee vahvistuvasti esille komponentin vah vistuessa, eli yrittäjälle erikoispuun saatavuus on yksi avainkysymyksistä. K 4: (Markkinat) Yrittäjätyypille markkinoiden olemassaolo ennen tuotannon aloittamista on tär keää. K 5: (Kehittämisprojektit) Erikoispuunkäytön kehittäminen nousee voimakkaasti esille. Yrittäjä tyyppi on halukas kehittämään erikoispuun käyt töä, mutta muutoksia yrityksessä ei kuitenkaan ole suunnitteilla. Seuraavassa taulukossa (taulukko 10) on esitetty erikoispuulajien kiinnostavuus yrityskoon mukaan tarkasteltuna. Kiinnostuneimpia laajentamaan tuotantoaan eri koispuulajeihin ja myöskin aloittamaan sen ovat pienyritykset. Kaikkiaan 70 % yrityksistä on kiin nostunut laajentamaan tuotevalikoimaansa erikois puihin. Kiinnostus laskee selvästi siirryttäessä suu riin yrityksiin, mutta osa niistäkin on kiinnostunut erikoispuulajeista, lähinnä lehtikuusesta. Taulukossa 11 on esitetty yrittäjiä kiinnostavat puulajit. Selvästi yrittäjiä kiinnostavimmat puulajit ovat haapa ja leppä; näistä puulajeista on kiinnostunut yli puolet kaikista yrittäjistä. Laadukas mänty ja koivu sekä lehtikuusi muodostavat toisen selkeän ryhmän, joista yrittäjät ovat kiinnostuneet. Sen si jaan laadukas kuusi, visakoivu, kataja, pihlaja sekä jalot lehtipuut eivät kiinnostaneet kuin muutamaa prosenttia vastanneista yrittäjistä. Tarvitun erikois puun määrä on noin 1000 m 3 vuodessa yritystä kohti, joka on huomattava ottaen huomioon yritys ten tämänhetkisen kapasiteetin. Yrittäjiltä kysyttiin, uskovatko he löytyvän mark kinoita edellä mainituille erikoispuutuotteille. 29 %:lta yrittäjistä oli kyselty jotakin erikoispuutuo Taulukko 10. Kyselyyn vastanneiden yritysten myön teisten vastausten %-osuudet puun käyttömäärän mu kaisesti luokiteltuna kysyttäessä kiinnostusta erikoispuun tuotantoon. Taulukko 1 1. Kaikkia vastanneita yrittäjiä kiinnostavat erikoispuulajit sekä 22 yrittäjän arviot tarvitsemastaan erikoispuun määrästä (pyöreä puu). tetta sekä 4 %:lla vastanneista yrityksistä oli jo sopimus asiakkaan kanssa jostakin erikoispuutuot teesta. Kysytyimpiä tuotteita ovat olleet leppä paneeli (14 yritystä) sekä haapainen laudepuu (11 yritystä). Kysymykseen, kuinka haluaisitte kehit tää erikoispuun hankintaanne, vastattiin taulukon 12 mukaisesti. Neljännes kyselyyn vastanneista yrittäjistä hank kii puunsa omana hankintana. Toinen merkittävä hankintatapa on metsänomistajien suoraan yrittäjäl le toimittama puu. Muita hankintatapoja tällä hetkel lä ovat hankinta suuremman puunhankkijan kautta sekä verkostotoiminta yritysten kesken. Joka kym menes yrittäjä hankkii puunsa puupörssin kautta. Tulevaisuudessa puuta haluttaisiin hankkia pää asiassa omana hankintana sekä metsänomistajien Puun käyttömäärä (m 3 /v) Kiinnostunut laajentamaan, % Valmis aloittamaan, % Alle 100 IA 70 101-1 000 73 69 1 001-10 000 78 83 10 001-100 000 50 50 Yli 100 000 33 66 Puulaji Osuus vastaajista % Määrä (m 3 /v) Laadukas mänty 14 4715 Laadukas kuusi 2 2 000 Laadukas koivu 12 1 500 Visakoivu 2 - Lehtikuusi 10 3 180 Leppä 29 5 000 Haapa 25 5 500 Kataja 2 - Pihlaja 2 - Jalot lehtipuut 2 - Yhteensä 100 21 895 Kärki Sahauskelpoisen erikoispuun laatuvaatimukset... 45 Taulukko 12. Kyselyyn vastanneiden yritysten nykyi set puunhankintatavat ja kiinnostus kehittää erikoispuun hankintaa tulevaisuudessa. Taulukko 13. Erikoispuutavaralajeille asetettavat vä himmäisläpimitta- ja pituusvaatimukset. Puulaji Vähimmäislpm (cm) Vähimmäispituus (m) Laatuluokka ABC ABC Lehtikuusi 25 14 8 2,40 2,00 2,00 Lehtipuut 20 14 8 2,40 2,40 2,00 suoraan toimittamana. Puupörssin välityksellä ha luaisi puukauppansa hoitaa joka kymmenes yrittä jä. Toisaalta kolmannes yrittäjistä on valmis käyt tämään mitä tahansa toimivaa puunhankintatapaa. 3.3 Laatuluokat Laatu- ja mittavaatimukset lehtikuuselle sekä lehti puulajeille (haapa, leppä, pihlaja, saarni, tammi, vaahtera, lehmus, jalava) muodostettiin yrityskyse lyn perusteella. Sen sijaan laadukkaan männyn, kuu sen ja koivun laatu- ja mittavaatimukset noudatta vat jo nykyisiä käytössä olevia vaatimuksia. Visa koivulle sekä katajalle ei aseteta tämän tutkimuk sen perusteella vaatimuksia läpimitan eikä pituu den suhteen, vaan oletetaan raaka-aineen olevan jalostettavissa olevaa puuta (rangat ja haarapöl kyt). Pihlajan tyvipölkkyjen vähimmäispituusvaa timus on 50 cm. Yrityskyselyn tuloksena esitetään kolmiluokkaista laatu- ja mittavaatimusluokittelua (taulukko 13). Laatuluokka a Tervettä puuta, jolla on erinomaiset puulajille omi naiset ominaisuudet. Puuaine tervettä, oksatonta tai melkein oksatonta ja suoraa. Puuaines saa oma ta merkityksettömiä vikoja, jotka eivät haittaa puun jalostamista ja käyttöä. Käyttö pääasiassa huone kalupuuksi, sorvattavaksi tai laadukkaaksi puuse pänpuuksi. Laatuluokka b Normaalilaatuista puuta, jonka puuaineessa on seu raavia vikaisuuksia yksi tai useampia: - lievää lenkoutta - terveitä oksia, läpimitaltaan pieniä tai keskikokoisia - vähäinen määrä pieniläpimittaisia kuivia oksia - lievästi epämuotoinen runko Käyttötarkoitus yleensä sahatuksi puutavaraksi, pa neeliksi tai käyttöesineisiin. Laatuluokka c Puu, joka ei vikaisuutensa takia kelpaa laatuluok kiin A tai B. Puu on kuitenkin suurimmaksi osaksi teollisesti käytettävissä. Rungossa ja puuaineessa esiintyy seuraavia vikoja: - voimakasta lenkoutta - haaraisuuttaja voimakkaita poikaoksia - väri vikoja - oleellisia hyönteisvahinkoja Puun pääasiallinen käyttötarkoitus on käyttö- ja koriste-esineiden valmistus. 4 Tulosten tarkastelu Yrityskyselyn perusteella on suhtautuminen erikois puun käyttöön Savo-Kaijalan alueella positiivisen kiinnostunutta. Negatiivista palautetta kyselyyn ei tullut, vaikkakin vastausprosentti jäi alhaiseksi. Hankintatapa Haluaisin kehittää (%) Hankin nyt (%) Omana hankintana 24 25 Metsänomistajien suoraan toimittamana 21 22 Ns. kyläterminaalien kautta 3 - Suuremman puunhankkijan kautta 6 14 Toimimalla verkostona muiden yritysten kanssa 6 17 Puupörssin välityksellä 12 8 Kaikki em. tavat 28 8 Pienet puunhankintafirmat - 3 Kokonaishankintaketju - 3 Yhteensä 100 100 Tutkimusartikkeli Metsätieteen aikakauskirja I /1997 46 Kyselyn perusteella kiinnostus erikoispuulajien ja lostukseen on vastanneiden yrittäjien keskuudessa voimakas, sillä 70 % vastanneista yrityksistä on kiinnostunut laajentamaan tuotevalikoimaansa eri koispuutuotteisiin. Osittainen selitys suhteellisesti suureen kiinnostuneiden määrään lienee se, että yrittäjät, jotka eivät ole erikoispuun käytöstä kiin nostuneita, eivät vastanneet yhtä innokkaasti kyse lyyn. Näin ollen kiinnostuneiden prosenttiosuus muodostuu suureksi. Jos oletetaan, että kyselyyn vastasivat kaikki erikoispuun käytöstä kiinnostu neet yrittäjät Savo-Karjalan alueella, muodostuu koko populaation estimaatiksi 18 % (vastauspro sentti 0.25 x kiinnostusprosentti kyselyssä 0.70). Eli tämän mukaan noin joka viides Savo-Karjalan alueen saha-ja puusepänyrityksistä olisi kiinnostu nut laajentamaan tuotevalikoimaansa erikoispuu tuotteisiin. Todellinen kiinnostuneiden prosentti luku löytynee näiden ääripäiden välistä. Ylivoimaisesti kiinnostavimmat puulajit ovat lep pä sekä haapa. Osa yrittäjistä on kiinnostunut ja lostamaan laadukasta mäntyä ja koivua sekä lehti kuusta. Kiinnostus jalojen lehtipuiden jalostami seen on hyvin vähäinen. Kyselyyn vastanneet yritykset hankkivat puunsa tällä hetkellä omana hankintanaan tai metsänomis tajan suoraan toimittamana. Samalta pohjalta halu taan puunhankintaa kehittää myös jatkossa. Puu pörssin kautta haluaa toimintaansa kehittää 12 % puunjalostajista sekä lisäksi ne yrittäjät (28 %), jotka ovat kiinnostuneet mistä tahansa toimivasta puunhankintasysteemistä. Koko tutkimusaineiston kattavalla pääkompo nenttianalyysilla haettiin tekijöitä, jotka ratkaise vat yrittäjien halukkuuden käyttää erikoispuuta raa ka-aineenaan. Analyysin tuloksena saatiin yksi koko tutkimuspopulaation käyttäytymistä kuvaava kom ponentti ja neljä erilaisia tekijöitä kuvaavia kom ponentteja (uusi tuotantomahdollisuus, erikoispuun hankinta, markkinat, kehittämisprojektit). Osa yrittäjistä pitää erikoispuunjalostusta uutena tuotantomahdollisuutena. Erikoispuun jalostuksesta saatetaan hakea tasoittajaa ja tuotannon monipuo listajaa suhdanteiden vaihdellessa. Erikoispuunhankinta nousee tuotannon alkaessa tai sitä lisättäessä ongelmaksi. Tähän vaikuttanee erikoispuun epävarma saatavuus, epätietoisuus laa tu- ja mittavaatimuksista sekä mahdollisesti tuo tantoon tarvittavan raaka-aineen pienet ja vaikeasti ennakoitavat määrät. Vakiintumaton erikoispuiden ostohintataso on myös omiaan käristämään eri koispuunhankinnan ongelmia. Tässä nousevat esil le myös erilaiset vaihtoehtoiset hankintamuodot (puupörssi, yritysverkostojen kautta hankittava puu), joihin kyselyn perusteella on kohtalaista kiinnos tusta. Erikoispuujalosteille löytyvät markkinat ovat ana lyysin mukaan yksi merkittävä tekijä. Kyselyssä kävi ilmi, että kolmannekselta yrittäjistä oli kyselty erikoispuutuotteita. Tilanteeseen saattaa vaikuttaa vielä se, että sahaajat ja sahatavaran käyttäjät eivät välttämättä kohtaa toisiaan markkinoilla, eivätkä tiedä toistensa tarpeista. Analyysin neljäs komponentti toi esiin kehittä misprojektit, joita kaivataan erikoispuun käytön edistämiseksi (koulutus, neuvonta, konsultointi). Tämä tarkoittaa logistista ketjua alkaen aina asiak kaan tarpeista jatkuen tuotekehitykseen, tuotantoon, varastointiin ja kuljetukseen sekä sivutuotteiden käsittelyyn. Tähän liittyen erikoispuulajien pätevät laatu- ja mittavaatimukset ovat ensisijaisen tärkeät (Erikoispuiden käytön lisäämisprojektiin ..., 1995). Kuten taulukosta 6 kävi ilmi, kyselyyn vastan neet yrittäjät ovat esittäneet mitta vaatimuksensa lehtikuusen, haavan ja lepän osalla normaalilaatui selle tai sitä heikompilaatuiselle puuainekselle ja rungolle. Muutamat olemassa olevat lähteet tuke vat osaltaan tutkimustuloksia. Verkasalon (1993 a) tutkimus tukee yrittäjien asettamia mittavaatimuk sia lehtikuusen osalla; samoin Lounan ja Valkosen (1995) tutkimuksessa lehtipuille asetetut mittavaa timukset ovat yrittäjien lehtipuille asettamien vaa timusten kanssa yhteneviä. Keinäsen ja Tahvanai sen (1995) Pohjolan Jalot Puut -opas sen sijaan antaa minimiläpimittojen osalta huomattavasti poik keavia arvoja. Sama on huomattavissa pituusvaati muksissa lehtikuuselle ja haavalle; muille puulajeille pituusvaatimukset ovat samaa luokkaa. Kuvassa 3 on esitetty logistinen ketju erikois puun laatu- ja mittavaatimusten määrittämisestä. Käyttäjä-orientoituneessa mallissa on lähtökohta na määrittää laatu-ja mittavaatimukset loppukäyttö tasolta. Kuluttajan (lopullisen asiakkaan tai jatko jalostajan) mieltymykset ja tarpeet ohjaavat voi makkaasti tarvittavalle raaka-aineelle asetettuja kri teerejä. Kuluttajan (lopputuote) vaatimuksia ohjaa Kärki Sahauskelpoisen erikoispuun laatuvaatimukset... 47 Kuva 3. Käyttäjä-orientoitunut malli erikoispuun laatu vaatimusten määrittämisestä. vat esimerkiksi uudet toteuttamiskelpoiset ideat ja trendit suunnittelussa. Näin syntyneen kysynnän kautta ohjataan puun jatkojalostajia (lopputyöstöt), jotka haluavat tarjota kuluttajilleen oikeanlaista ta varaa ja vaativat sitä myös sahayrityksiltä (alku työstö). Ketju etenee puunhankintaan, jonka kautta on hankittava sopivaa raaka-ainetta jalostukseen. Mah dollisia hankintamuotoja erikoispuun osalla ovat puupörssi, terminaalikauppa, metsäpalveluyrittäjät sekä normaalit hankinta- ja pystykaupat sekä kä teiskaupat. Tässä vaiheessa on oltava tiedossa yh teiset laatu- ja mittakriteerit, joiden mukaisesti puu voidaan apteerata. Sopivan erikoispuun kasvatus on metsäyrittäjän harteilla, joka viime kädessä pys tyy vaikuttamaan huomattavasti optimaalisen raa ka-aineen tuottamiseen. Kaikkiaan erikoispuun kohdalla sekä puun ky syntä että tarjonta on vielä melko vähäistä. Tärkeää on kuitenkin, että viesti erikoispuun käytöstä ja markkinoiden mahdollisuuksista saadaan liikkeel le. Toimiva erikoispuutavaran hankintasysteemi sekä jokapäiväiseen käyttöön saadut laatu- ja mitta vaatimukset ovat parhaita keinoja saada erikois puulajien hankintakustannukset kilpailukykyiselle tasolle. Tällöin on luotu toimintamahdollisuudet järkevälle erikoispuuta jalostavalle pienteollisuu delle. Kirjallisuus Ahola, V.K. 1952. Mukurapuista. Communicationes In stituti Forestalls Fenniae 40(18). 10 s. Alestalo, A. & Hentola, Y. 1967. Leppä sulfaattikeitos sa. Paperi ja Puu 50: 25-27. Björklund, T. & Ferm, A. 1982. Pienikokoisen koivun ja lepän ominaisuudet. Folia Forestalia 500. 37 s. Biihl, A. & Zöfel, P. 1994. SPSS fiir Windows Version 6. Addison-Wesley GmbH, Bonn. 503 s. Eisele, F-L. 1989 a. Überlegungen zur Sortierung, Ver messung und Kennzeichnung von Rundholz (I). Holz- Zentralblatt 124: 1921-1923. 1989 b. Überlegungen zur Sortierung, Vermessung und Kennzeichnung von Rund holz (II). Holz-Zentralblatt 126: 1969-1971. Erikoispuiden käytön lisäämisprojektiin liittyvä markkina selvitys. 1995. Erikoispuiden käytön lisäämis -projekti. Kuopion yliopisto, koulutus-ja kehittämisk. 12 s. + liitt. Grammel, R. 1989. Forstbenutzung; Technologie, Ver wertung und Verwendung des Holzes. Pareys Studien texte 67, Hamburg. 193 s. Hakkila, P. & Winter, A. 1973. On the properties of larch wood in Finland. Communicationes Instituti Forestalls Fenniae 79(7). 45 s. Holmäsen, I. 1991. Pohjolan puut ja pensaat. WSOY, Helsinki. 177 s. Kallio, M. & Salo, S. 1992. Puupörssi. Helsingin kauppa korkeakoulun julkaisuja D-156. 44 s. Keinänen, E. & Tahvanainen, V. 1995. Pohjolan jalot puut. Pohjois-Savon erikoispuiden käytön lisäämis projekti. 160 s. Knigge, W. & Schulz, H. 1966. Grundriss der Forst benutzung. Verlag Paul Parey, Göttingen. 584 s. Kuula, M., Kallio, M., Salo, S. & Vepsäläinen, A. 1992. Puupörssi-simulaattori käyttöohje. Helsingin kauppa korkeakoulun julkaisuja D-158. 20 s. Kärkkäinen, M. 1980. Suomalainen haapa- ja poppeli lajeja (Populus) koskeva kirjallisuus 1759... 1979. Silva Fennica 14(4): 369-383. 1981. Haapa-ja poppelilajien (Populus) käyttö. Silva Fennica 15(2): 156-179. & Salmi, J. 1978. Tutkimuksia haapatukkien mittauk sesta ja teknisistä ominaisuuksista. Folia Forestalia 355. 45 s. Lakio, L.A. 1953. Puutavaralajit ja niiden valmistus. Keskusmetsäseura Tapio. Helsinki. 44 s. Lehtonen, 1., Pekkala, O. & Uusvaara, 0. 1978. Tervalepän (Alnus glutinosa (L.) Gaertn.) ja raidan (Salix caprea L.) puu- ja massateknisiä ominaisuuksia. Folia Forestalia 344. 19 s. Tutkimusartikkeli Metsätieteen aikakauskirja I /1997 48 Louna, T. & Valkonen, S. 1995. Kotimaisen raaka-aineen asema lehtipuiden teollisessa käytössä. Metsäntutki muslaitoksen tiedonantoja 553. 38 s. Luttinen, M. 1994. Puunhankinnan yhteistyömalli/Puu pörssi. Ylä-Savon instituutti, Sonkajärvi. 18 s. Mali, J. 1980. Kotimaisten puulajien ja tuontipuulajien tekniset ominaisuudet ja käyttö. Valtion teknillinen tutkimuskeskus, puulaboratorio. Tiedonanto 3. 43 s. Metsätilastollinen vuosikirja 1992. 1993. Metsäntutki muslaitos. 317 s. Opdahl, H. 1989. Avsmaling og volum hos osp (Populus tremula L.) i Sor-Norge. (Tapering and volume of aspen (Populus tremula L.) in South Norway. Med delelser fra Norsk institutt for Skogforskning. 43(2): 1-42. 1991. Bonitet, vekst og produksjon hos osp (Populus tremula L.) i Sor-Norge. (Site-index, growth and yield in aspen (Populus tremula L.) stands in South Nor way.) Meddelelser fra Norsk institutt for Skogforsk ning. 44(11): 1-44. Palen, M. 1994. Lapin erikoispuuprojektin toimintasuun nitelma. Metsähallituksen Koillis- Lapin hoitoalue, Kemijärvi. 10 s. Puu-Suomi -projekti. 1994. Puu-Suomi -projektin Puun käytön teemaryhmä. Toimintasuunnitelma. 7 s. Ranta, E., Rita, H. & Kouki, J. 1992. Biometria. Yliopisto paino, Helsinki. 569 s. Rohholzsortierungsbestimmungen fur den Gebrauch im bayerischen Staatswald. 1993. Staatsministerium fiir Ernährung, Landwirtschaft und Forsten, Munchen. 32 s. Saarnijoki, S. 1961. On muutakin visaa kuin koivun visaa! Metsätaloudellinen aikakauslehti 78: 257-259. Saarnio, R. 1976. Viljeltyjen visakoivikoiden laatu ja kehitys Etelä-Suomessa. Folia Forestalia 263. 28 s. Sairanen, P. 1982. Lehtikuusen ominaisuudet ja käyttö Neuvostoliiton mekaanisessa metsäteollisuudessa. Metsäntutkimuslaitoksen tiedonantoja 72. 25 s. Salmi, J. 1972. Suomalaisia ja ulkolaisia puulajeja. Osa I: Havupuut. Helsingin yliopisto, metsäteknologian laitos. Tiedonantoja 17. 227 s. —1977. Suomalaisia ja ulkolaisia puulajeja. Osa II: Lehtipuut A-N. Helsingin yliopisto, metsäteknologian laitos. Tiedonantoja 35. 282 s. —1978. Suomalaisia ja ulkolaisia puulajeja. Osa III: Lehtipuut O-O. Helsingin yliopisto, metsäteknologian laitos. Tiedonantoja 38. 298 s. Schalin, I. 1966. Harmaalepän merkityksestä käytännön metsätaloudessa. Metsätaloudellinen aikakauslehti 83(9): 362-366. Steuer, R.W. 1979. Die Rohholzsortierung und die Giite klassen. Holz-Zentralblatt 86: 1261- 1263. Tikka, P.S. 1954. Haapametsiköiden rakenteesta ja laa dusta. I. Rakenne. Communicationes Instituti Fores talls Fenniae 43. 33 s. 1956. Haapametsiköiden rakenteesta ja laadusta. 11. Laatu. Communicationes Instituti Forestalis Fenniae 45. 54 s. Tuimala, A. 1993. Lehtikuusipuun ominaisuudet ja käyt tö. Metsäntutkimuslaitoksen julkaisuja 464: 79-90. Uusvaara, O. & Pekkala, O. 1979. Eräiden ulkomaisten ja kotimaisten puulajien puu- ja massateknisiä omi naisuuksia. Communicationes Instituti Forestalis Fen niae 96(2). 59 s. Valkonen, S., Rantala, S. & Sipilä A. 1995. Jalojen lehtipuiden ja tervalepän viljely ja kasvattaminen. Metsäntutkimuslaitoksen tiedonantoja 575. 112 s. Vuokila, Y. 1960. Siperialaisten lehtikuusikoiden kehi tyksestä ja merkityksestä suomalaiselle metsätalou delle. Communicationes Instituti Forestalis Fenniae 52(5). 103 s. Verkasalo, E. 1990. Koivuja haapa mekaanisen metsä teollisuuden raaka-aineena Yhdysvalloissa. Metsän tutkimuslaitoksen tiedonantoja 367. 83 s. 1993 a. Lehtikuusen mittaus ja kauppa. Metsäntutki muslaitoksen tiedonantoja 464: 91-98. 1993 b. Lehtikuusitukkien laatu Suomessa. Folia Fo restalia 823. 27 s. Vepsäläinen, A. & Kuula, M. 1992. Puupörssin toteutta minen ja vaikutukset puukaupan logistisiin palvelui hin. Helsingin kauppakorkeakoulun julkaisuja D-157. 29 s. 47 viitettä SPECIES, FURNITURE TYPE, AND MARKET FACTORS INFLUENCING FURNITURE SALES IN SOUTHERN GERMANY TIMO KÄRKI The author is a Research Scientist, Finnish Forest Research Institute, P.O. BOX 68, FIN-80101 Joensua Finland, Fax +358 13 251 4111, E-mail Timo.Karki® metla.fi 2 Kärki ABSTRACT The aim of this study was to investigate the furniture markets in Central Europe as a case study focusing on the economic area of Munich, southern Germany. The topics studied were the breakdown by tree species and furniture types and the marketing factors used to differentiate the product or the sales approach. The factors were divided into 19 different classes, which were classified empirically during the study: price, design, tree species, furniture selection, domestic product, general quality, durability of materials, „green attributes", extra equipment, delivery in pieces, new products, finishing (waxed and oiled), waxing, oiling, stable construction, guarantee, material, comfort, and practicality. Data were collected from the advertisements in provincial and local newspapers and from advertisements distributed by furniture stores. The most important marketing factor was design (28.2 %), which was almost 10 percentage points more than for the second most important factor, species (18.7 %). Price was ranked third (14.9 %). „Green attributes" and finishing (waxed and oiled), which are ecological factors, were ranked fifth and seventh. „Green attributes" were even more common among the marketing factors than the concept of quality. According to this study, there are large differences in the marketing of wood furniture in different European countries, e.g. Finland and Germany. In Finland, furniture is marketed as good value for the money, large selection, and domestically produced. In Germany, design, species and pro environmental factors are highlighted. When „green attributes" and species are used in advertising, pro-environmental thinking is being emphasized. P&P Statement: The results of this study highlight the use of different species in the German furniture markets and the pro-environmental thinking of furniture advertising. This information has practical importance for furniture manufacturers and exporters. Species, furniture type, and market factors influencing ... 3 SPECIES, FURNITURE TYPE, AND MARKET FACTORS INFLUENCING FURNITURE SALES IN SOUTHERN GERMANY INTRODUCTION Furniture plays an important role in satisfying the social needs of the family. For example, there are differences among family members in their use of the living-room; the children of the family may use it very little (when not watching TV) and adult use is mostly in the evenings and on weekends. Home is not only a physical state but it also includes the net of social relations for family members, friends, and neighbors. Home also indicates the identity of the person, reflecting his or her life style and personal values. People do not want a home that is too perfect, but rather a home of the style which can be changed when the demands of the owner change. Furniture represents the changeable part of the home, and with its help the individual is able to express his / her personality and feelings. Moreover, through furniture the family can also express their social status. Furniture is commonly associated with memories and experiences of childhood, as most people remember the furniture in their childhood home. In general, furniture plays a very stable and essential role in our lives; and of the objects in a home, for most people, furniture is the most important (18). When the concept of ..furniture" is considered, the importance of style and fashion must be clarified. According to Trankle (31), style is understood to be longer lasting than fashion. Furniture can also be seen as a status symbol. The furniture style symbolizes the „style" and the living standards of the family. The importance of a living room is even greater because most of the social communication with guests takes place there. The consumer's behavior in marketing situations has been investigated in many studies (4, 6, 8, 12, 15, 17, 19, 20, 23-26, 29, 32, 33). In Schreiber's study (26), five criteria are listed that must be taken into consideration when the buying behavior of a consumer is studied: 1) price-consciousness, 2) quality-consciousness, 3) original, unusual decision, 4) planned decision; and 5) nonconformist decision. On the one hand, the price-consciousness of the consumer varies according to how urgently he / she needs the product. On the other hand, discounts in prices very obviously affect customers positively. One factor in the buying process is that consumers may have chosen in advance the style they are looking for (e.g., Scandinavian style). This reduces the need to sort through many sources of information and 4 Kärki markets already in the beginning of the search. Another factor is the style of the consumer's previous furniture. Households that were satisfied with their old furniture need a shorter time for consideration and they simplify their decision making by considering fewer alternatives than those households that were not satisfied with their old furniture. In the buying process, the wife plays an important role because it is usually she who provides the impetus to change the furniture. The husband has a greater influence on the price group of the furniture and the wife on factors such as style, product name, and store (5, 27). Despite these differences, most buying decisions are made jointly. The children also have an important role in buying furniture; according Bucker's study (4), in most cases (66 %) the opinions of children are asked when furniture is bought for the living room. It was also shown that buyers with higher education and better incomes use more sources of information when buying furniture. The choice of furniture store from the standpoint of price has been investigated in two studies (7, 13). Other relevant factors are product aspects (quality, variation of assortments) and convenience of shopping (location, parking possibilities, client information, etc.). Different payment possibilities (installment, loan, cash discount) may also influence the buying decision. Other aspects affecting furniture purchases are material, form, color, standards, and certificates. Already at the beginning of their search, consumers have problems deciding which information sources they will use in the decision process. Sources of information about furniture available for sale are furniture magazines, ads, shop windows, etc. These are considered in Fig. 1. Figure 1. The household in the information field (22). The basics in marketing and advertising research have been reported in several studies (16, 22, 28). The interpretation of advertising has also been investigated in many studies (1, 2, 3, 9, 10, 17). In order to study green advertising, Soikkeli (30) explored furniture ads from decorating magazines. Haiminen (14) concluded that, in general, green marketing ads are honest and relevant; but they transmit too little environmental information. Altogether, ad Species, furniture type, and market factors influencing 5 studies are rarely used in investigating forest product markets. Therefore, the aim of the present research was to investigate the furniture markets in Central Europe as a case study, focusing on the economic area of Munich, and to determine how these markets are influenced by species, furniture type, and marketing factors, i.e. how the products are sold. MATERIAL AND METHODS The material used in this study consisted of ads collected in southern Germany in the economic area of Munich (6 Mill, inhabitants). Data were collected from ads in provincial and local newspapers and from the ads of furniture stores. Two newspapers were checked systematically (Tageszeitung Munchen TZ and Freisinger Kreisbote). Ads were collected from 24 different furniture stores, including both small and large firms. Data were collected during the period from week 40/1997 until week 40/1998 and the ads were collected every second week. Marketing factors were divided into 19 different classes, which were classified empirically during the study: price, design, species, selection, domestic product, quality, durability of the materials, „green attributes", extra equipment, delivery in pieces, new product, finishing (waxed and oiled), waxing, oiling, stable construction, guarantee, material, comfort, and practicality. For example, „green attributes" meant that the concept „pro-environment" was highlighted. In the advertisement, it could be said that the timber used in the furniture came from sustainably managed forests. The data comprised 3246 pieces of furniture, of which 2347 were wooden. The furniture was divided into four main types: solid wood, partly solid wood, composite, and upholstered furniture. In this study, furniture that was to any extent upholstered was not included in the solid wood furniture category, even if the construction of the furniture frame was solid wood. Instead, this kind of furniture was considered to be upholstered furniture, which is not considered in this article (Table 1). Composite furniture is a piece of furniture that is built from composite panels that are overlaid with veneers or printed-grain plastic materials. The data were analyzed by cross-tabulation. In order to study the marketing factors used for different types of structure, a Chi-Square-Test was also used, but no significant difference was found. This was an expected result while it is supposed that certain arguments, like price, have a greater significance in solid and plate-structured furniture than in partly solid. Kärki 6 TABLE 1. Summary of the key characteristics of the material studied. Almost all ads were obtained from free advertisement material delivered by mail. In the newspapers, there were few ads to analyze. The furniture stores were categorized as large chain stores or local companies. More than 80 percent of the furniture ads were from firms that operated locally. Large multinational furniture companies were intentionally omitted while only the market situation in the southern Germany was studied. Almost half of the furniture was made from solid wood and 20 percent was partly solid. More than 30 percent of the furniture was composite. Smaller, environmentally oriented („green") firms clearly specialized in certain types of advertisements. This difference was not only obvious in terms of the number of furniture pieces advertised but also in the marketing arguments used to sell them; e.g., the ads of environmentally oriented firms differed considerably from those of the larger firms. The larger firms attempted to fill out their ads with colors and numbers, while the ads of „green" firms were more subdued. In these latter ads, there were nature photos, the text was restrained, and the prices were given inconspicuously. For the same size of ad paper, „green" firms had fewer items than the larger companies did. The ads of environmentally oriented firms were easy to read, and the appearance of the ad demonstrated what kind of product was for sale: Characteristics Frequency Percentage (n = 2347) Furniture Frequency Percentage (n = 2347) Type of magazine Provincial newsp. 15 0.7 Wardrobe 524 22.3 Local newsp. 9 0.4 Secretary 339 14.4 Advertisement 2323 98.9 Bookshelf 203 8.6 Chair 183 7.8 Type of store Coffee table 143 6.1 Chainstore 370 15.8 Dining table 132 5.6 Local 1977 84.2 Bed 128 5.5 Writing table 109 4.6 Type of furniture Bedroom furn. 98 4.2 Solid 1133 48.3 Kitchen furn. 89 3.8 Partly solid 468 19.9 TV-table 72 3.1 Composite 746 31.8 Shelf 70 3.0 Glass cabinet 62 2.6 Use Glass cabinet Outdoor 142 6.1 / closet 38 1.6 Indoor 2205 93.9 Garden table 36 1.5 Table group 33 1.4 Bench 32 1.4 Bedside table 31 1.3 Trunk 11 0.5 Couch 8 0.4 Couch group 3 0.3 Rocking chair 2 0.1 Species, furniture type, and market factors influencing ... 7 a close-to-nature and good-quality product. The good sense and feelings of the customer were appealed to in these ads. In this study the furniture was divided into 22 categories. The largest groups were wardrobes (22.3 %) and secretaries (14.4 %). Groups like bookshelves, chairs, coffee tables, dining tables, beds and writing tables also exceeded 100 units. Bedroom sets (bed, wardrobe, and bedside table) and kitchen furniture sets (five kitchen cupboards without electrical equipment) were also included in the study. RESULTS TABLE 2. Summary of marketing factors and species. When all results were considered, design was the most frequently mentioned marketing factor (Table 2). The proportion of the marketing arguments based on design was 28.2 %, which was almost 10 percentage points higher than for the second most important argument: species. Price was ranked third. Interestingly, „green attributes" and finishing (waxed and oiled), which are ecological attributes, placed fifth and seventh. „Green attributes" were even more common among the marketing factors than the concept of quality. Marketing factor Frequency Percentage (n = 2347) Tree species Frequency Percentage (n = 2039) Design 661 28.2 Pine 675 33.1 Tree species 440 18.7 Beech 514 25.2 Price 350 14.9 Spruce 272 13.3 Practicality 235 10.0 Alder 203 10.0 „Green Oak 118 5.8 attributes" 169 7.2 Maple 60 2.9 Quality 139 5.9 Birch 35 1.7 Finishing 115 4.8 Linden 26 1.3 Selection 63 2.7 Cherry 25 1.2 Extra equip. 58 2.5 Teak 25 1.2 Durability Ash 20 1.0 of materials 57 2.4 Rattan 16 0.8 Stable constr. 20 0.9 Mahogany 12 0.6 Comfort 20 0.9 Walnut 11 0.5 New product 10 0.4 Root tree 10 0.5 Material 5 0.2 Acacia 10 0.5 Guarantee 5 0.2 Others 7 0.4 8 Kärki Pine dominated as a furniture species, making up one-third of the construction material. Beech took second place. The next two species, spruce and alder, lagged far behind. The proportions of tropical species like teak, mahogany, and acacia were very small. TABLE 3. Origin of the furniture and colors used in painted furniture. The origin of the furniture was also studied (Table 3). In most of the ads, however, the origin of the furniture was not mentioned; in 149 ads the origin was given. According to this, the origin was mentioned as Scandinavian in half of the mentions of origin and German in one-third of them. Some of the solid wood furniture was painted a color that was intended to match that of a particular species. Beech was imitated in 90 cases and pine in 45 cases. It should be noted that some of the composite pieces of furniture that were overlaid with plastic were lower in price. Of the 658 plate-structured pieces of furniture in this study, 51.7 percent were plastic overlaid (Table 4). TABLE 4. Number of composite furniture pieces, by species, percentage overlaid with wood veneer, and percentage overlaid with printed-grain plastic (n = 658). Origin Frequency Percentage (n = 149) Colour Frequency Percentage (n = 197) Scandinavian 81 54.4 Beech 90 45.7 German 53 35.6 Pine 45 22.8 Other 15 10.0 Cherry 19 9.6 Alder 14 7.1 Chestnut 12 6.1 Other 17 8.7 Species No. of Percentage overlaid Percentage composite pieces with wood veneer overlaid with plastic Beech 384 42.4 57.6 Alder 79 43.4 56.6 Maple 50 46.0 54.0 Oak 39 89.7 10.3 Ash 19 42.1 57.9 Birch 11 54.5 45.5 Cherry 11 90.7 9.1 Walnut 11 0 100 Root tree 10 0 100 Other 44 66.7 33.3 Species, furniture type, and market factors influencing ... 9 The proportion of composite furniture overlaid with printed-grain plastic was about the same in the three largest groups of species (beech, alder, and maple). For some species, such as oak and cherry, the proportion was much lower due to the fact that oak and cherry are considered expensive species for furniture and they are seldom imitated. TABLE 5. Distribution of furniture by species and type (n = 2039). Table 5 shows that softwood furniture (pine and spruce) was most often solid wood. All tropical hardwood furniture (teak, mahogany, and acacia) was solid wood. Alder and oak had quite equal proportions of every structural type of furniture, but beech furniture was mostly composite and only occasionally solid. The use of different species by category, and prices according to category, type, and species are shown in Tables 6 and 7, respectively. Tree species Total Solid Partly solid Plate structured Pine 675 85.6 13.0 1.4 Beech 514 4.5 20.8 74.7 Spruce 272 68.4 31.3 0.3 Alder 203 28.6 32.5 38.9 Oak 118 21.2 45.8 33.0 Maple 60 - 16.7 83.3 Birch 35 8.6 60.0 31.4 Linden 26 92.3 - 7.7 Cherry 25 56.0 - 44.0 Teak 25 100.0 - Ash 20 - 5.0 95.0 Rattan 16 100.0 - - Mahogany 12 100.0 - - Walnut 11 - - 100.0 Root tree 10 - - 100.0 Acacia 10 100.0 - - Others 7 42.9 14.2 42.9 10 Kärki TABLE 6. Proportion (%) of furniture by species and category (n =1651). TABLE 7. Variation in furniture prices (DEM)* according to category, type, and species. Tree species Wardrobe Secretary Bookshelf Chear Coffee table Dining table Bed Writing table Pine 34.0 33.7 11.7 45.7 18.1 32.0 58.4 42.4 Beech 23.8 22.0 33.0 8.7 36.2 28.4 31.0 42.4 Spruce 22.0 16.3 6.1 7.8 8.5 6.6 1.8 8.6 Alder 9.8 8.9 19.8 2.4 18.1 10.7 6.2 1.1 Oak 2.9 5.2 6.6 7.0 6.1 11.5 - 1.1 Maple 1.2 5.8 6.6 1.6 1.1 1.6 1.8 3.3 Birch 1.2 1.8 5.6 1.1 - - - Linden 0.6 1.5 - - 6.4 1.6 - Cherry 1.2 0.3 4.6 0.8 1.1 2.5 1.1 Teak 0.3 10.6 1.1 0.9 Ash 0.4 1.5 - - - Rattan - - 10.6 0.9 Mahogany 0.6 1.0 2.4 1.1 0.9 - - Walnut 0.3 1.0 - 1.1 0.9 - Root tree 0.3 1.0 1.5 - - Acacia 0.2 0.9 2.4 0.9 Others 2.1 1.2 3.0 Total 491 326 197 170 125 125 120 97 pes Type of furniture Solid Pine Spruce Alder Oak Linden Partly solid Beech Spruce Pine Alder Oak Composite Beech Alder Maple Oak Ash Wardrobe 1405 1386 2067 4157 898 815 829 902 1073 1399 679 726 1110 1238 279 Mean 1459 974 705 Secretary 715 776 722 1537 1078 693 514 620 700 430 314 171 588 299 101 Mean 747 657 354 Bookshelf 2050 2903 5165 849 1942 610 1498 2465 2704 1557 676 1234 464 Mean 2700 2438 1524 Chair 206 294 213 295 156 149 - 174 - 112 - 139 Mean 230 261 119 Coffee table 550 360 781 998 494 270 - 598 249 49 164 128 301 Mean 633 331 193 Bed 493 98 1249 - - 678 275 564 - 609 445 775 Mean 512 734 579 Mean species 889 1130 1229 1317 675 898 724 829 1408 1273 719 604 886 623 152 Mean total 1047 899 579 a Deutsch mark Species, furniture type, and market factors influencing ... 11 According to Table 7, the price relations between solid, partly solid, and composite furniture seem to be similar for all six categories of furniture. Solid-structured furniture was the most expensive, partly solid was in the middle price range, and composite was the least expensive. For chairs and beds, however, the average prices for partly solid furniture were even higher than for solid-structured. Moreover, softwoods were very obviously used only for solid and partly solid pieces of furniture, and hardwoods were the most commonly used species in composite furniture. Certain price groups could be identified. On the whole, oak furniture was the most expensive price group, closely followed by alder. Furniture from beech had noticeably lower prices when made into partly solid furniture but considerably higher prices as composite pieces, compared to those from oak or alder. Pine and spruce furniture was the least expensive in solid and partly solid furniture. Composite maple and beech furniture was considerably more expensive than that from alder or oak. DISCUSSION The material in this study consisted mainly of the advertisement material for 24 furniture stores. The largest stores advertised all kinds of furniture equally. Interestingly, the domestic origin of the furniture was not considered to be an advantage. Softwoods were sold as solid-structured furniture. The proportions of these species in composite furniture were marginal. The tropical hardwoods (teak, mahogany, and acacia) were sold only in solid-structured form. Another very obvious aspect was that the softwoods were used only for solid and partly solid pieces of furniture and hardwoods were the most common species in composite furniture. This is obviously a consequence of the price difference between hardwoods and softwoods, since the average import price for hardwoods was considered to be higher than that of softwoods (11). According to this study, with regard to species, there were certain price groups of furniture. On the whole, oak furniture was the most expensive with alder following as a close second. Furniture from beech was noticeably less expensive when used as partly solid but considerably more expensive as composite furniture. Pine and spruce furniture was the least expensive in solid and partly solid furniture. Composite maple and beech furniture was considerably more expensive than that made from alder and oak. The most important source of error in this material was the subjective sampling of the 12 Kärki advertisements, even though all kinds of furniture stores were included in the material. Along with the geographical concentration of the sampled material (economic area of Munich), this permits only limited possibilities to apply the results elsewhere in Central Europe. Unfortunately, the furniture purchasers were not interviewed; interviews would have made possible detailed comparison between advertising arguments and purchaser requirements. Nevertheless, the material gives an indication of furniture advertising and markets in southern Germany. When comparing the results to other European countries, e.g. Finland, the marketing of wood furniture differs greatly. In Finland, the furniture is marketed as good value for money, large selection, and domestically produced (22); in Germany, design of the furniture, species, and pro-environmental aspects are highlighted. The difference in marketing between Finland and Germany is even greater for solid wood furniture. When „green values" and species are used in marketing, the advertiser wants to highlight pro-environmental thinking. The results of this research on the use of different species by Central European furniture manufacturers is important to the Scandinavian wood processing industry because Central Europe represents their most important export markets. In Scandinavia, silviculture has traditionally focused mainly on three species: pine, spruce, and birch. This study clearly showed that in furniture markets, other species, like alder, are also popular. This raises interesting questions: What will be the future of the silviculture of alder ? Are there any economic possibilities in Scandinavian forestry to plant and manage alder? Species, furniture type, and market factors influencing ... 13 LITERATURE CITED 1. Baacke, D. & Kubler, H-D. eds. 1989. Qualitative Media Research. Max Niemeyer Verlag, Tubingen. 365 pp. (in German). 2. Behrens, K.C. 1963. Sales Advertising. Publishing House Dr. Th. Gabler, Wiesbaden. 227 pp. (in German). 3. Bender, M. 1976. The Measurement of Advertising Success in Advertising. Physica- Verlag, Wurzburg. 212 pp. (in German). 4. Bucker, H.P. 1986. Furnishing Determinants in Private Households. Studienverlag Dr. N. Brockmeyer, Bochum. 265 pp. (in German). 5. Davis, H.L. 1970. Dimensions of material Roles in consumer decision making. J. of Marketing Res. 3/1970: 168. 6. Dichter, E. 1964. Handbook of Consumer Motivations. The Psychology of the World of Objects. McCraw-Hill Book Co., New York. 486 pp. 7. Diller, H. 1977. Price as a quality indicator. Die Betriebswirtschaft 37/1977: 219. (in German). 8. Engell, J.F. and R.D. Blackwell. 1982. Consumer Behavior. The Dryden Press, Minsdale 111. 690 pp. 9. Felser, P. 1991. The intensity of advertisement research of big advertisers. Ph.D. thesis, Faculty of Economics, University of Freiburg/Switzerland. 245 pp. (in German). 10. Fischerkoesen, H.M. 1967. Experimental Advertisement Success Prognoses. Publishing House Dr. Th. Gabler, Wiesbaden. 181 pp. (in German). 11. Foreign trade according to Goods and Countries, year 1996. 1997. Statistiscal Office of Germany, Wiesbaden. 135 pp. (in German). 12. Foxell, G.F. 1980. Consumer Behaviour. A Practical Guide. Groom Helm Ltd, London. 207 pp. 13. French, N.D., J.J. Williams and W.A. Chace. 1972. A shopping experiment on price quality relationship. J. of Retailing 48/1972: 3. 14. Haiminen, H. 1997. Green advertising in furniture business: A comparison between consumer and professional advertising. M. Sc. Thesis. Univ. of Helsinki, Finland. 104 PP -15. Hermanns, A. 1979. Consumer and Advertisement Success. Publishing House Gieseking, Bielefeld and Köln. 335 pp. (in German). 16. Huttner, M. 1965. The Basics of Marketing Research. Publishing House Dr. Th. Gabler GmbH, Wiesbaden. 360 pp. (in German). 17. Jacobi, H. 1963. Advertisement Psychology. Publishing House Dr. Th. Gabler, Wiesbaden. 140 pp. (in German). 18. Jyläskoski, B. 1993. Advertising as a semiotic communication transmitter. M. Sc Thesis. Univ. of Vaasa. 113 pp. (in Finnish). 19. Langholz - Leymore, Varda. 1975. Hidden Myth: Structure and Symbolism in Advertising. Heinemann Educational Books Ltd., London. 208 pp. 20. Leven, W. 1991. Glance Behavior of Consumers. Physica-Verlag, Heidelberg. 286 pp. (in German). 21. Meyer, P.W. 1986. Advertisement and Advertising Research. Univ. of Augsburg. 170 pp. (in German). 22. Pakarinen, T. and K. Turunen. 1999. The success factors of wooden furniture and furniture manufacturing. Res. Notes 88. Univ. of Joensuu, Faculty of Forestry, Finland. 68 pp. (in Finnish). 14 Kärki 23. Peter, P.J. and J.C. Olsson. 1987. Consumer Behaviour. Marketing Strategy Perspectives. Richard D. Irwin Inc., Homewood 111. 698 pp. 24. Petermann, G. 1963. Market Position and Market Behavior of a Consumer. Publishing House Dr. Th. Gabler, Wiesbaden. 80 pp. (in German). 25. Schiffman, L.G. and L.L. Kanuk. 1978. Consumer Behaviour. Prentice Hall, Engelwood Cliffs, N.J. 489 pp. 26. Schreiber, K. 1965. Purchasing Behavior of a Consumer. Publishing House Dr. Th. Gabler, Wiesbaden. 176 pp. (in German). 27. Scott, R.A. 1970. Husband - wife interaction in a household. Southern J. of Business 5/1970: 218. 28. Seufert, R. 1966. Advertisement Science - The Theory and Practise of an Advertisement. Volume 1 and 2. C.E. Poeschel Verlag, Stuttgart. 1702 pp. (in German). 29. Sinclair, S.A. & Hansen, B.G. 1993. The relationship between purchase decisions and quality assessment of office furniture. Wood and Fiber Science 25(2): 142-152. 30. Soikkeli, M. 1997. Green advertising in furniture business. M. Sc. Thesis. Univ. of Helsinki, Finland. 91 pp. 31. Tränkle, M. 1972. The Living Culture and Manners. Tubingen. 175 pp. (in German). 32. Vinson, D.E. and J.E. Scott. 1977. The role of personal values in marketing and consumer behaviour. J. of Marketing (4): 44-49. 33. Wiswede, G. 1973. Motivation and Consumer Behavior. Ernst Reinhardt Verlag, Munchen. 211 pp. (in German). Diameter distribution, stem volume and stem quality models for grey alder (Alnus incana) in eastern Finland TIMO KÄRKI 1 , MATTI MALTAMO 2 and KALLE EERIKÄINEN 1 I Faculty of Forestry, University of Joensuu, FIN-80101 Joensuu, Finland; 2 Finnish Forest Research Institute, Joensuu Research Station, FIN-80101 Joensuu, Finland Received 22 December 1999; accepted 14 January 2000 Key words: Alnus incana, taper curve, stem-rot, Weibull distribution Abstract Stand-level predictions of basal-area diameter distributions, height, positions of lowest dead and live branches and stem-rot are provided for grey alder in eastern Finland. The usability of models was tested by calculating the accuracy of predicted volume characteristics. The data were collected from 33 stands situated in the provinces of North Karelia and North Savo in eastern Finland. These stands were regenerated naturally, but some have been man aged. One to three angle-count sample plots were placed systematically in each stand using a 1 m 2 ha -1 basal-area factor. The models can be applied in two settings: tree diameters are either measured or not. The prediction of branch height characteristics, especially height to the lowest dead branch, proved problematic due to weak correlations between these characteristics and other tree dimensions. Compared to previous studies it was found that stem-rot was higher in grey alder than in spruce but lower than in white birch. Introduction During a period of shifting cultivation in the 18th and 19th centuries, grey alder (Alnus incana) spread throughout southern Finland. In silvicultural cuttings after World War O, alder gave way to other tree species and alder dominated forests have since been on the decrease. The greatest concentration of grey alder is in eastern Finland where shifting cultivation continued longer. According to the Eighth National Forest Inventory of Finland, in the provinces of North Karelia and North Savo three percent of the volume of growing stock consists of grey alder (Metsätilastollinen vuosikirja 1996). Its proportion of the number of trees per hectare is as high as 17%; this is in contrast to the average of 9.6% for southern Finland. Grey alder has been little used as a raw material for the forest industry. However, the importance of alder species has recently clearly increased. The increasing need for inexpensive energy has focused attention on renewable 2 biomass resources. In Finland, researchers have suggested that the woody biomass produced by fast-growing deciduous species, such as grey alder to be used as an alternative to imported energy sources (Saarsalmi et al. 1985, 1991, 1992). Recently, alder has also become a realistic alternative for the reforestation of former agricultural areas. Alder species are actinorrhizal plants; they live in symbiosis with nitrogen-fixing actinomycetes (.Frankia spp.). Therefore, they have the ability to increase the nitrogen content of the soil. The profitability of growing grey alder depends also on its suitability as a raw material for industries. The wood is homogenous and easy to work with, making it suitable for various products and many special purposes, such as for making music instruments (Kärki 1997b). When considering the quality of grey alder, it has been found that dimensions, decay and knottiness are the most important factors influencing the value of sawn alder timber (Kärki 1997). According to Kärki (1999 a), rot and discoloration in grey alder wood cause 13 percent of the decrease in the quality of individual pieces of lumber. In a living tree, rot normally begins from damage to the tree; damage can be caused by management operations, animals, and environmental damage such as storm breakage, which can initiate entry for fungal infection (Kallio 1972). Information about grey alder resources in Finland is available mainly from the national forest inventories. This information is useful for large-scale esti mates, but there is a lack of knowledge and models for predicting the amount and quality of alder in stands and forest holdings. The first growth and yield studies on grey alder were made before World War II by Miettinen (1933). However, when considering models, only taper curve have been estimated in Finnish conditions (Mäkinen 1984). This taper model concerns both grey and black alder and is applicable only for small trees. The mean tree diameter in Mäkinen (1984) was close to 4 cm and the largest diameter was nearly 15 cm. In Finland, the actions of forest management planning are based on inventory by compartments, where only mean stand characteristics, such as mean diameter and height, basal area and age, are measured. Calculations of stand characteristics are then based on theoretical diameter distribution, tree height and taper models (Laasasenaho 1982; Veltheim 1987; Kilkki et al. 1989). Using this approach it is possible to determine stand volume quite precisely but the wood quality cannot be estimated. For example, the calcula tion of the amount of sawn timber is based only on tree dimensions and thus the amounts of different quality classes of sawn wood cannot be calculated. There are several models which include height of the lowest living branch and lowest dead branch and stem rot characteristics for main tree species in Finland (Koivunen 1989; Mäkelä 1990; Rouvinen et al. 1997). In most cases these models include variables which are not normally measured as 3 mean stand characteristics and cannot therefore be applied directly in forest management planning. In some recent studies it has been proposed that stem quality should be added to field measurements for forest management planning (Rouvinen et ai. 1997). In the study of Rouvinen et ai. (1997), in addition to normal measure ments of mean diameter and height, mean height of the lowest living branch and lowest dead branch were measured. Firstly, branch height character istics were generalized to all trees of the stand using measured mean branch height characteristics to calibrate existing branch height models by stand. It was expected that these branch heights determine the timber quality classes between branchless logs, logs with dead branches and logs with living branches. Finally, these branch heights and tree taper were used in determin ing the optimal volumes of logs of these timber quality classes (Rouvinen et ai. 1997; Puumalainen 1998). The aim of this study was to estimate models of diameter distribution, stem volume and stem height for grey alder. These models allow us to derive forest level estimates for grey alder using mean stand characteristics. In addi tion, the characteristics needed to establish models describing tree quality characteristics were determined. Finally, various stand volume characteristics are calculated using the new models and independent tree measurements. Material and methods Data collection The material in this study, included 33 stands (minimum area of 0.5 ha) situ ated in the provinces of North Karelia and North Savo in eastern Finland (Figure 1). The stands were naturally regenerated, but at least part of them have been under management. From one to three angle-count sample plots were located systematically in each stand de'pending on the size of the stand using a relascope with a basal area factor (BAF) of 1m 2 ha -1 . A total of 75 sample plots were measured. Tree species and diameter of each tree were recorded. Stand age, stand basal area, forest site type, basal area median diameter, median and dominant heights were measured on the sample plots. Mean stand characteristics are shown in Table 1. The stand characteristics are the same as those used in conventional compartment level forest inventories. Tree height and height of the lowest living branch and lowest dead branch were measured on the shortest, tallest and median tree of each plot (Table 2). These sample trees, altogether 230, were felled and the diameter and bark thickness at relative heights of 1, 5, 10, 20, 30, 40, 50, 60, 70, 80 and 90% were measured. If rot was detected, the diameter at stump height and the 4 Figure 1. The material for this study includes 33 stands situated in the provinces of North Karelia and North Savo in eastern Finland. length of rot were registered. The stems were also cut into short pieces in order to detect the rot anywhere else in a stem that could have started from a broken branch. The average characteristics for rot diameter at stump height (mm) and rot length (m) are presented in Table 3. Weibull distribution Because the trees were chosen by angle-count sampling, the diameter distri bution was weighted by basal area and therefore called basal area diameter distribution (e.g. Maltamo et al. 1995). Basal area diameter distributions have been mainly used in Finland (e.g. Kilkki and Päivinen 1986; see also Gove 5 Table 1. Mean stand characteristics of the sample plot data. Table 2. Mean characteristics of the sample tree data. and Patil 1998). In the study of Kilkki and Päivinen (1986), it was stated that a small number of measured trees from angle-count sample plot can be used to estimate parameters for the Weibull-distribution. However, a more reliable description of tree stock is obtained if there are several angle-count sample plots in one stand and they are combined to estimate the parameters of the Weibull-distribution (Maltamo et al. 1995; Maltamo 1997). Sample plots were combined for each stand to get a more reliable basal area diameter distribution. Consequently, there were an average of 40 sample Table 3. Summary of rot statistics. Variable Mean sd Min Max d gM 13.3 2.890 7.8 19.6 hgM 13.1 2.705 5.9 18.3 h dom 13.9 2.271 8.0 18.0 T 31.8 6.521 16.0 50.0 G 17.8 6.112 6.0 36.0 Explanation of the variable codes: sd = standard deviation. Min = minimum value of variable. Max = maximum value of variable, dg m = median tree diameter at breast height, cm; hg M = median of tree heights, m; T = mean stand age, yr, hjom - average dominant height, m; G = basal area, m 2 ha -1 . Variable Mean sd Min Max Msmm 12.8 4.366 3.8 25.8 m 12.7 3.010 2.6 20.4 6.9 2.396 1.2 14.1 4.3 2.568 0.5 11.2 Explanation of the variable codes: sd - standard deviation of variable, Min - minimum value of variable, Max = maximum value of variable, d 1.3 -= diameter at breast height, cm; h = total tree height, m; h fib ~ height to the first living branch, m; h f db = height to the first dead branch, m. Variable Mean sd Min Max RDSH 51 LR t .82 32.08 10.00 150.00 1.87 2.88 1.00 16.50 Explanation of the variable codes: RDSH = rot diameter at the stump height, mm, LR = length of the rot, m. 6 trees per stand. Theoretical basal area diameter distributions were fitted using the two-parameter approach of the Weibull distribution. The probability density function of the two-parameter Weibull distribution is: where x is random variable (tree diameter), b is scale parameter, and c is shape parameter. Weibull parameters for each sample plot were estimated using maximum likelihood estimation by solving the natural logarithm of the likelihood func tion of the Weibull density function. After the Weibull distributions were estimated, their parameters were regressed using mean stand characteristics as explanatory variables. Height models Due to the assumed correlation between sample plots and trees in one stand, mixed models were used when tree height and heights to the lowest living branch and the lowest dead branch were modelled. In these models, the generalized least square (GLS) method was used. The GLS method divides the residual variation of the model into variances between stands, between plots in a stand and among trees in a plot. The covariance structure of the material is taken into account in the parameter estimates (Lappi 1986). The fixed part of the model consists of explanatory variables and the random part includes stand, plot and tree effects in this case. When the reliability of these models was determined, stand and plot effects were calculated using sample tree measurements and estimated residual variances (Searle 1971). These calculated effects were then used to calibrate the height models. Taper equation Stem volumes were calculated using tree taper function instead of volume equations. As a result, it was more flexible to calculate volume estimates for different proportions of the tree stem. There are various ways to determine tree taper mathematically. In Finland, the most common taper estimation approach is the so called polynomial taper curve method. Although, the poly nomial taper curve method has mainly been used and studied on most of the common Finnish tree species (Scotch pine, Norway spruce, birch), it has also been applied successfully on other tree species elsewhere (Määttä 1987; Heinonen et ai. 1996). For example, in a study on Eucalyptus cloeziana (F. Muell.), Eerikäinen et ai. (1999) found the polynomial taper curve method to be the most accurate stem diameter and volume estimator when compared f{x) = §(§)' 1 exp (- (f ) c ), (0 1 yi variable Coefficient x -value Variable Coefficient x -value co 1.3767 c0 -0.2964 ci 0.2633 8.31 c\ 0.1809 5.12 cj -0.6713 8.14 c2 0.0006 7.93 c3 29.0155 5.01 % correct 72.61 66.96 X 2 -value 24.69 14.49 G 2 -value 290.43 307.89 Explanation of the variable codes: x 2 = value of x 2 -test; G 2 -value = Log Likelihood — test value. 16 Table 11. Regression coefficients for predicting the length of rot in alder stems. The prediction of volume characteristics using varying sample tree information The applicability of the height and taper-curve models to predict volumes of different parts of tree stems were tested using the same approach as in the study of Rouvinen et ai. (1997). However, lowest dead branch and lowest living branch were in most trees located so near each other in tree stem that it was not sensible to calculate estimates for part of the stem which includes dead branches. Instead, the branchless part below lowest dead branch was considered. The other characteristics which were chosen was part of the stem below lowest living branch which includes both branchless and dead branches sections. In the study of Rouvinen et ai. (1997), tree stem parts including branches of different quality were further divided into logs using cross cutting. In this study, total volumes of these parts were directly calculated using varying information. These calculations were done using the sample tree material of this study. Measured diameters at relative heights of the stem of sample trees were used in the cubic spline interpolation (Lahtinen and Laasasenaho 1979) to calculate estimates of the real stem volumes. When different height models were considered, tree diameters at breast heights were known. Tree and branch heights were predicted using the models in Tables 5 and 6, respectively. Different stem volumes were calculated using the constructed taper curve (Tables 7 and 8). To obtain more accurate volume predictions Model LR LR variable Coefficient T-value Coefficient T-value Intercept 4.5363 2.69 1.9296 2.50 Bo.OSh 0.4295 3.83 - - 1IG 15.0000 3.15 9.4892 2.41 Hdbl - -0.0027 -1.93 Lnihfdb) RDSH -0.8445 -3.02 0.0563 6.33 RMSE 1.68 1.54 d2 adj. 0.13 0.40 Explanation of the variable codes: LR = length of the rot, RDSH = rot diameter at the stump height, Bo.OSft = bark thickness at 5% height, G = basal area, h f ab " height of the first dead branch, = height of the dry branch limit. 17 Table 12. Reliability of different volume estimates (1-3) using Methods 1-4. of different parts of the stem, different calibrations of the presented models were tested. Calibration was done by utilizing stand and plot effects which were calculated using sample tree measurements and estimated residual variances of mixed models. In these calibrations, it was assumed that sample tree characteristics were measured from a median tree as mean stand characteristics. The methods used were: Method 1: No calibration. Method 2: Calibration of the total tree height model (Table 5). Method 3: Calibration of the models for total tree height (Table 5) and height to the first living branch (Table 6). Method 4: Calibration of the models for total tree height (Table 5), height to the first living branch and height to the first dead branch (Table 6). The results of the calculations are presented in Table 12. The accuracy of prediction of the total stem volume and volume to the first living branch were predicted almost as equal but in the case of volume to the lowest dead branch it was considerably worse. Calibration slightly improved the accuracy of all predictions. However, when only the model for total tree height was calibrated, the RMSE for prediction of the volume to the lowest dead branch was slightly increased. Volume RMSE RMSE-% BIAS BIAS-% Method 1 1 12.2 11.9 2.0 2.0 2 9.9 11.4 0.5 0.6 3 16.8 26.7 -2.5 -3.9 Method 2 1 9.9 9.6 1.5 1.4 2 8.4 9.6 <0.1 <0.1 3 17.0 26.9 -2.7 -4.3 Method 3 2 8.3 9.5 <0.1 <0.1 Method 4 3 13.2 22.1 0.8 1.4 Explanation of the variable codes: Volume 1 - estimated total volume; Volume 2 = estimated volume: stem volumes integrated from the ground to the first living branch; Volume 3 = estimated volume: stem volumes integrated from the ground to the first death branch; RMSE = Root mean square error of volume estimates, m 3 ; RMSE-% = Relative RMSE, %; Bias = Bias of volume estimates, m 3; Bias-% = Relative bias, %. 18 Discussion In this study, different models of tree and stand characteristics were constructed for grey alder. These models include diameter distribution, tree height, height to the first living and dead branch, taper curve and some rot models. The models can be applied basically with or without diameters. 1) When only mean stand characteristics are known. Diameter distribu tion is first predicted, and the theoretical tree diameters obtained from this distribution are used to predict total tree height and branch heights. Finally, volume characteristics can be determined using predicted tree diameter and height. In this situation, the average rot of the grey alder stock can be assessed using the information in Figure 3. However, in this study it was not possible to validate this procedure because there was no test material available for which the real stand structure was known. 2) When tree diameters and sample trees are measured, it is also possible to calibrate different height models and obtain more accurate information about these characteristics (Table 12). Correspondingly, the rot models presented in this study can also be applied to obtain more detailed information about amount of rot on the tree level. However, this requires some additional treewise measurements to be made. The construction of different models indicated that there was much variation in grey alder stands that cannot be explained. Diameter distribution, tree height, height to the lowest and tree taper were able to be predicted using mean stand characteristics but prediction of other characteristics required additional measurements. Models of tree rot presented in this study included variables which are not normally measured in field measurements of forest management planning. This means that the applicability of these models is mainly on situations where some specific inventories are carried out. For example, stand inventory concentrating on tree quality or sawing production planning in certain grey alder stands could utilise models presented here. The description of diameter distribution using the two-parameter approach of Weibull distribution prediction proved to be accurate in terms of x 2 statis tics. An alternative would have been to use percentile based distributions (Borders et al. 1987). In the study of Knowe et al. (1997), stand structure and dynamics of young plantations of red alder were considered. To produce current and future distributions, they compared different methods, including diameter distribution prediction and stand table projection by percentiles. Prediction of branch-height characteristics, especially height to the first dead branch, proved problematic, due to the poor correlation between these characteristics and other tree dimensions. Also the calibration of these models using sample tree measurements only slightly improved the applicability of 19 these models. This has also been found for other tree species (Rouvinen et ai. 1997; Uusitalo and Kivinen 1998). However, branch-height charac teristics are important when tree quality is considered. Rouvinen et ai. (1997) proposed that average branch-height characteristics should also be measured in field inventories when other stand characteristics are collected. These variables could then be generalized directly for all trees of the stand. This was also tested in this study but the results were not good. Using non-parametric kernel-smoothing, Uusitalo and Kivinen (1998) developed a system for describing bivariate tree diameter and height to the first dead branch. The benefits of this kind of approach occur especially in situations where the correlation structures between certain characteristics are complex (Silverman 1986). However, these non-parametric methods require measure ment of several sample trees per stand and could not be used for the material in this study. When the reliability of the polynomial taper curve for grey alder (Equation 5) was tested in a previous study by Kärki et al. (1999b), the estimated relative bias and the standard error of stem volume estimates were about 0.4% and 9.0%, respectively (Kärki et al. 1999b). Laasasenaho (1982) found that when the taper polynomial (based on Fibonacci series) was used, the estimated relative standard error of the volume estimates varied from 7.2 to 8.5% for pine, spruce and birch. A good starting point for analysis of reliability is to compare volume estimates calculated with a taper polynomial to those estimated using a volume function. A simple nonlinear volume function for grey alder has been presented earlier (Kärki et al. 1999b) using the same data as used in this study for the taper polynomial. These two methods gave almost the same estimates of total stem volume. The taper polynomial was a more reliable and stable estimator, especially in the largest diameter classes. Rot is more common in grey alder than in spruce which are common associates in eastern Finland. Mäkelä (1990) studied the probability of rot in Norway spruce stems. If we compare the results of these studies, some interesting differences can be seen. The proportion of rotten trees according to age classes was different from grey alder in these studies; in Norway spruce stands, the proportion of rotten trees was smallest in the beginning of the rotation time, while in alder, stands the proportion was smallest in the middle of the rotation time. The difference in alder stands could be explained as being due to the way alder regenerates (sprouting). There is another obvious difference in the diameter classes; the proportion of rotten Norway spruces is smallest in large diameter classes while in alder stands the proportion of rotten trees is highest in large diameter classes. Norokorpi (1979) found that when du and the age of the tree increase, the length of the rot in spruce stems shows a linear increase. In that study, the 20 volume of the trees and the volume of the rot were also compared and found to be exponential. Tamminen (1985) has studied butt-rot in standing spruces. The variables described the form and growth of the stem of the sample trees (diameter increment at 1.3 m height in last 10 years). The success rate in the prediction model was 68%, which is comparable to the results of the present study. Mäkelä (1990) also obtained similar results for spruce; the success rate for prediction of rot in a tree was 60%. According to these studies, the success rates (73 and 67%) found in this study can be considered good. When comparing the proportion of rotten trees at stump height in different age classes in grey alder and white birch (Betula pubescens) (Verkasalo 1997), the proportion of rotten grey alder stems in different age classes is significantly lower than in white birch. A recent study of white birch (Verkasalo 1997) found that for trees of seed origin the proportion of trees with rot was 32, 37 and 45 percent for trees under 40 years, 41 to 60 years and over 60 years of age, respectively. The proportions were higher for trees of sprout origin for the same age classes, 45, 53 and 61 percent. The grey alder in the current study had 36, 29 and 33 percent trees with rot in the same age classes. This is a significant result while it is normally thought that the rot in grey alder is much more common than in other hardwood species in Finland. Conclusions The material used was not optimal for this type of study. Trees were chosen using a relascope, and the number of sample trees per plot was low. All measurements were made on temporary sample plots. The low number of sample trees affected the possibility to test different models together. However, in the earlier data on grey alder, which were based mainly on the national forest inventory, the number of sampled trees in pure stands of grey alder was very low. Earlier information about grey alder has come mainly from single measurements of grey alder trees in mixed stands. Consequently, this study and the constructed models reveal new and important information about pure stands of grey alder in eastern Finland. Acknowledgements This study was carried out at the Faculty of Forestry of the University of Joensuu. We thank Prof. T. Pukkala and Prof. E. Verkasalo for valuable comments on the manuscript and Dr. J. von Weissenberg for suggestions on English editing. 21 References Borders, 8.E., Souter, R.A., Bailey, R.L. and Ware, K.D. 1987. Percentile-based distributions characterize forest stand tables. Forest Science 19: 97-104. Cox, D.R. 1989. The Analysis of Binary Data. Chapman and Hall, London. 236 pp. Eerikäinen, K.P.A., Mabvurira, D. and Saramäki, J. 1999. Alternative taper curve estimation methods for Eucalyptus cloeziana (F. Muell.). Southern African Forestry Journal 184: 12-24. Gove, J.H. and Patil, G.P. 1998. Modeling basal area-size distribution of forest stands: A compatible approach. Forest Science 44: 285-297. Heinonen, J., Saramäki, J. and Sekeli, P.M. 1996. A polynomial taper curve function for Zambian exotic tree plantations. Journal of Tropical Forest Science 8: 339-354. Hosmer, D.W. 1989. Applied Logistic Regression. Wiley, New York. 307 pp. Kallio, T. 1972. An example on the economic loss caused by decay in growing spruce timber in South Finland. Silva Fennica 6: 116-124 (in Finnish with English summary). Kärki, T. 1997. Haapa- ja leppätukkien kysyntä, hankinta ja laatu. Joensuun yliopisto, metsätieteellinen tiedekunta. Tiedonantoja 53. 78 pp. (in Finnish). Kärki, T. 1999 a. Predicting the value of grey alder logs (Alnus incana) based on external quality. Silva Fennica 33: 13-23. Kärki, T., Eerikäinen, K., Heinonen, J. and Korhonen, K.T. 1999 b. Harmaalepän (Alnus incana ) tilavuustaulukoL Metsätieteen aikauskirja - Folia Forestalia 1/1999: 39—49 (in Finnish). Kilkki, P. and Päivinen, R. 1986. Weibull function in the estimation of the basal area DBH distribution. Silva Fennica 20: 149-156. Knowe, S.A., Ahrens, G.R. and Deßell, D.S. 1997. Comparison of diameter-distribution - prediction, stand-table-projection, and individual-tree growth modeling approaches for young red alder plantations. Forest Ecology and Management 96: 207-216. Koivunen, J. 1989. Männyn ja koivun kuivien ja elävien oksien alkamiskorkeuden ennustam inen. Metsänarvioimistieteen pro gradu-tutkielma. Helsingin yliopisto. 44 p. (in Finnish). Laasasenaho, J. 1982. Taper curve and volume functions for pine, spruce and birch. Commu nicationes Instituti Forestalls Fenniae 95: 1-63. Lahtinen, A. and Laasasenaho, J. 1979. On the construction of taper curves by using spline functions. Communicationes Instituti Forestalls Fenniae 95: 1-63. Lappi, J. 1986. Mixed Linear Models for Analyzing and Predicting Stem Form Variation of Scots Pine. Communicationes Instituti Forestalls Fenniae 134. 69 pp. Määttä, M. 1987. Taper Curve Functions for Cupressus lusitanica and Pinus pat ula in North-East Tanzanian Softwood Plantations. University of Helsinki, Department of Forest Mensuration and Management. 11 pp. + appendices. Mäkelä, H. 1990. Tyvilahon esiintymistodennäköisyys etelärannikon kuusissa ja lahon korkeuden ennustaminen. Metsänarvioimistieteen pro gradu-tutkielma, Joensuun yliop isto. 51 pp. (in Finnish). Mäkinen, T. 1984. Harmaalepän ja tervalepän runkomuoto, kuorimallit ja runkokäyrämallit. Metsänarvioimistieteen pro gradu-tutkielma, Helsingin yliopisto. 60 pp. (in Finnish). Maltamo, M. 1997. Comparing basal area diameter distributions estimated by tree species and for the entire growing stock in a mixed stand. Silva Fennica 31: 53-65. Maltamo, M., Puumalainen, J. and Päivinen, R. 1995. Comparison of Beta and Weibull func tions for modelling basal area diameter distribution in stands of Pinus sylvestris and Picea abies. Scandinavian Journal of Forest Research 10: 284-295. 22 Metsätilastollinen vuosikirja 1996. Statistical yearbook of forestry. Metsäntutkimuslaitos. Maa-ja metsätalous 1996: 3 (Helsinki, in Finnish). Miettinen, L. 1933. Tutkimuksia harmaalepiköiden kasvusta. Communicationes Instituti Forestalls Fenniae 18.1: 1-100 (in Finnish with German summary). Norokorpi, Y. 1979. Old Spruce Stands, Amount of Decay and Decay-Causing Microbes in Northern Finland. Communicationes Instituti Forestalls Fenniae 97. 77 pp. Press, W.H., Teukolsky, S.A., Vetterling, W.T. and Flannery, B.P. 1992. Numerical Recipes in Fortran. The Art of Scientific Computing, 2nd edn. University Press, Cambridge. 963 pp. Puumalainen, J. 1998. Optimal cross-cutting and sensitivity analysis for various log dimen sion constraints by using dynamic programming approach. Scandinavian Journal of Forest Research 13: 74-82. Reed, D.D. and Green, EJ. 1984. Compatible stem taper and volume ratio equations. Forest Science 30: 977-990. Rennolls, K., Geary, N. and Rollinson, TJ.D. 1985. Characterizing diameter distributions by the use of Weibull distribution. Forestry 56: 57-66. Rouvinen, S., Kangas, A. and Maltamo, M. 1997. Männikön laatujakauman kuvaaminen oksa rajatiedon avulla kuvioittaisessa arvioinnissa. Metsätieteen aikakauskirja-Folia Forestalia 4/1997: 477-492 (in Finnish). Saarsalmi, A., Palmgren, K. and Levula, T. 1985. Leppäviljelmän biomassan tuotos sekä ravinteiden ja veden käyttö. Folia Forestalia 628. 24 pp. (in Finnish). Saarsalmi, A., Palmgren, K. and Levula, T. 1991. Harmaalepän vesojen biomassan tuotos ja ravinteiden käyttö. Folia Forestalia 768. 25 pp. (in Finnish). Saarsalmi, A., Palmgren, K. and Levula, T. 1992. Harmaalepän ja rauduskoivun biomassan tuotos ja ravinteiden käyttö energiapuuviljelmällä. Folia Forestalia 797. 29 pp. (in Finnish). Searle, S.R. 1971. Linear Models. John Wiley, New York. Silverman, B.W. 1986. Density Estimation for Statistics and Data Analysis. Chapman & Hall. 175 pp. Tamminen, P. 1985. Butt Rot in Norway Spruce in Southern Finland. Communicationes Instituti Forestalis Fenniae 127. 52 pp. Uusitalo, J. 1995. Pre-Harvest Measurement of Pine Stands for Sawing Production Planning. University of Helsinki, Department of forest resource management. Publications 9.96 pp. Uusitalo, J. and Kivinen, V.-P. 1998. Constructing bivariate DBH/dead branch height distri bution of pines for use in sawing production planning. Scandinavian Journal of Forest Research 13: 510-515. Van Laar, A. 1985. Form quotients and stem form of Eucalyptus cloeziana. South African Forestry Journal 133: 18-23. Veltheim, T. 1987: Pituusmallit männylle, kuuselle ja koivulle. In: Mäkelä, H. and Salminen, H. (Eds) Metsän tilaa ja muutoksia kuvaavia puu-ja puustotunnusmalleja. The Finnish Forest Research Institute, Research Notes 398. 265 pp. (in Finnish). Verkasalo, E. 1997. Hieskoivun laatu vaneripuuna. Metsäntutkimuslaitoksen tiedonantoja 632. 483 pp. (in Finnish). Silva Fennica 33(1) research articles 13 Predicting the Value of Grey Alder {Alnus incana) Logs Based on External Quality Timo Kärki Kärki, T. 1999. Predicting the value of grey alder ( Alnus incana) logs based on external quality. Silva Fennica 33(1): 13-23. The quality of grey alder logs {Alnus incana ) was studied by sawing sample logs from two different forests in November 1995-February 1996. For grading of grey alder logs and sawn timber the proposed system of Keinänen and Tahvanainen (1995) plus the reject -grade was used. In general, grey alder logs have knots from the base to the top. All types of knots appear, and the length of the knot-free section is small at the base. In small-dimensioned logs there are fewer knots than in larger logs. Especially in large top logs, there were many more fresh knots than in other types of logs. Evidently, in different types of logs the different grades of sawn timber are located in comparable sections along the length. It also seems that the worse the grade class was, the longer was also the length of the class. The most common reasons for decreasing grade were dry knots and discoloration. Keywords Alder, log quality, sawn timber quality, log value Author's oddress University of Joensuu, Faculty of Forestry, P.O. BOX 111, FIN-80101 Joensuu, Finland Fax +358 13 251 4444 E-mail timo.karki@forest.joensuu.fi Received 17 October 1997 Accepted 9 November 1998 1 Introduction Grey alder has had restricted use in Finland and is approved only for minor uses such as fire wood. According to The Bth National Forest In ventory, in southern Finland the proportion of grey alder from the stem number varies from 0.9 % (Ahvenanmaa) to 18.2 % (Itä-Häme). The volume varies from 0.3 % (Lounais-Suomi) to 2.7 % (Pohjois-Savo). Of the alders growing in Finland, grey alder is more common than red alder (Alnus glutinosa) (Table 1). A significant proportion of red alder forests are planted com pared to grey alder forests, which are all natural. The technical properties and use of grey alder wood have been studied by Schalin 1966, Ales talo and Hentola 1967, Hakkila 1970, Lehtonen etc. 1978, Björklund and Ferm 1982 and Kärki 1997 a. The yield of grey alder have been studied by Miettinen 1933, Mäkinen 1984 and Valkonen etc. 1995. In Finland, alder timber is used very little. According to Louna and Valkonen (1995), in research articles Silva Fennica 33(1) 14 1993 about 20 000 m 3 of domestic alder timber was used. This is a rough estimate, which takes into account mainly the use of alder by large companies. The system suggested by Keinänen and Tah vanainen (1995 ) for grading grey alder logs and sawn timber (Tables 2 and 3) was used in this study for valuing logs and timber. Here this clas sification is also used to test the suitability. Several methods are used to study the internal quality of logs. The methodology and classifica tion based on external defects have been studied by Orver 1970, Hanks 1976, Weslien 1983, Flink Table 1. Amount of alder timber in southern Finland (with bark) according to The Bth National Forest Inventory (1000 cbm). Table 2. Quality classification of grey alder logs (Keinänen and Tahvanainen (1995) Table 3. Grading system for grey alder sawn timber, proposed by Keinänen and Tahvanainen (1995). Species Logs Industrial tree Slash Total Grey alder 52 1660 749 2461 Red alder 237 1071 71 1380 Total 289 2731 820 3841 Factor of dimension or quality A Grade B c Top diameter Min. 19 cm Min. 15 cm Min. 13 cm Min. length 21 dm 21 dm 21 dm other lengths in intervals of 1 dm Accuracy of length ± 3 cm ± 3 cm + 3 cm Ring width Even No special requests No meaning Crook 2 cm/m 2 cm/m 3 cm/m Number of knots per m / max. knot size - all knots 0 pes Max. 4 pes Max. 6 pes - fresh knots Not allowed Max. 2 pcs/3 cm Max. 3pcs/7cm - dry knots Not allowed Max. 2 pcs/3 cm Max. 3pcs/4cm - rotten knots Not allowed Not allowed Max. 2pcs/3cm Twist Not allowed Slightly allowed No meaning Cracks Not allowed Not allowed Not restricted Discoloration Not allowed Slightly allowed Max. 1/2 diam. Rot, centered Max. 1 cm Max. 5 cm Max. 1/2 diam. Factor of dimension Grade or quality A B C Min. width 180 mm 150 mm 130 mm Max. number of knots, on the worse side and worst meter 2 pes 3 pes 5 pes From the number of knots can be, pcs/mm - fresh knots 2 pes/ 10 mm 2 pcs/40 mm 3 pcs/60 mm - dry knots Not allowed 2 pcs/30 mm 3 pcs/40 mm - rotten knots Not allowed Not allowed 1 pcs/30 mm - bark knots Not allowed 1 pcs/30 mm 1 pcs/40 mm Cracks, over 100 mm length Not allowed Max. length 300 mm Not restricted Growth Even Not restricted Not restricted Tension wood, % of the length Not allowed Max. 10 % Max. 20 % Discoloration, % of the length Not allowed Max. 10 % Max. 15 % Rot centered, width Not allowed Max. 20 mm Max. 50 mm Kärki Predicting the Value of Grey Alder [Alnus incana) Logs Based on External Qualify 15 man 1985, Blomqvist and Nylinder 1988, Klinkhachorn et al. 1988, Harless et al. 1991 and Grace 1993. Uusitalo 1994 studied the predic tion of sawn wood quality in pine stands. Ver kasalo (1995) has listed possible reasearch meth ods for studying logs: 1. Traditional test sawing and grading of individual pieces of sawn timber 2. Test sawing used by Technical Research Center of Finland (VTT) 3. Describing the internal quality by X-ray and NMR techniques 4. Describing the internal quality after rotary-cutting 5. Theoretical sawing (simulation) In traditional test sawing, the logs are bucked and sawn according to the commercial patterns. The defects are measured or classified on the surfaces of the pieces of sawn timber, and the pieces are valued according to dimension, grade and price per unit volume. This method is used in the present study. In the VTT method, the logs are sawn into thin slices and the geometry of the slices and the position, quality and size of the knots are measured on a coordination table. Ac cording to this information, the stem or the log can be described mathematically with knots in XYZ-coordinates and sawn theoretically in giv en ways. They can then be graded and valued according to the dimensions, grades and prices per unit volume. In X-ray (also called Computer Tomography CT) and NMR-techniques the internal quality of the log can be described by measuring the changes in timber, which can mean internal knots and other defects. An advanced method for describ ing the internal quality of the log is rotary-cut ting, after which defects in the logs can be ob served and measured from veneer by manual or automatized scanning methods. Theoretical saw ing based on dimensions and external quality of the logs is one more option; these have been studied, for example, by Hallock and Galiger 1971, Richards 1979, Nakata 1986, Liljeblad et al. 1988 and Meimban 1991. The aim of this study was to investigate the potential to predict the value of grey alder logs by external properties. This is divided into the following subaims: 1. Predicting the value of logs according to external properties. 2. Reasons for variation in quality of logs and sawn timber. 3. Quality distribution of sawn timber in logs of different type/grade. 4. Testing the suitability of the quality classification of Keinänen and Tahvanainen (1995). 2 Material and Methods The research material comprised 229 grey alder logs from two pure grey alder stands, both situ ated in Northern Savo, one in Maaninka and the other in Nilsiä (Table 4). The trees were ran domly selected with the limitation that the logs had to be sawable (stem form). Total log volume was 18.1 m 3. The logs were sawn during No vember 1995-February 1996. For grading grey alder logs and sawn timber the proposed system of Keinänen and Tahvanainen (1995) plus the reject -grade was used. Top diameter (with bark) of the individual logs ranged from 112 to 295 mm. Most of the logs were in the top diameter classes of 13 cm to 18 cm. In the first stand (Maaninka) the logs were smaller than in the second stand (Nilsiä). Table 4. Test logs divided into 1 cm top diameter classes. Top diameter class l Stand 2 Total 11 4 1 5 12 8 4 12 13 20 4 24 14 28 7 35 15 17 13 30 16 18 12 30 17 21 10 21 18 17 8 25 19 6 7 13 20 6 8 14 21 2 1 3 22 - 4 4 23 1 2 3 24+ 1 1 2 Total 149 82 229 research articles Silva Fennica 33(1) 16 Table 5. Minimum, maximum and average values of sawlog data. The average values for sawlogs are presented in Table 5. Volume of sawlogs were calculated according to the formula: where Prior to sawing, the dimensions, shape, branchi ness and other visible defects were measured at given intervals 0-0.5 m, 0.5-1.5 m, 1.5-2.5 m, 2.5-3.5 m and 3.5+ m. These factors needed to be measured to be able to classify the logs into different quality classes (Table 2). After this, the logs were sawn unedged (i.e. by flat slicing). This "saw-dry-rip" -method is commonly used when hardwood logs are sawn, in which the logs are first sawn unedged and the sawn timber (with bark) is then dried. Edging of sawn wood is per formed after drying. The test logs were sawn into one dimension, 19 mm. After sawing, the sawn wood was graded according to the requirements listed in Table 3. The values for the sawn pieces and logs were calculated in the following way: 1 The sawn pieces were graded according to quality classifications (Table 3). The minimum length of one grade in a piece of lumber was 40 cm. This was estimated to be the minimum length of a component. Table 6. Prices of grey alder sawn timber. 2 The volume for each quality in a sawn piece was calculated. 3 According to quality and volume, the value for each sawn piece was calculated. The values used for different quality classes are shown in Table 6. The value for other wood material (by-products incl. bark), that could be used as firewood or material for fibre- or chipboards, was 150 FIM. 4 The values of individual sawn pieces and by-prod ucts in the log were used to calculate the value for the whole log. These log values were used in mod els. The data were analyzed with SPSS software. An equation was devised by regression analysis to depict the relationship between the log value and the external properties of the log. The relation ship between the grade of a piece of sawn timber and the location of that piece in the log was investigated by tabulation. The distribution of the knots in the different parts of the logs was investigated by cross-tabulation. The logs were grouped into four log grades in order to give at least 50 observations for each grade, but this procedure was not always successful. 3 Results 3.1 Quality of Logs There were fewer fresh knots in small-dimen sioned butt and top logs than in larger logs. Especially in large top logs, there were consider ably more fresh knots than in other types of logs. In all types of logs, fresh knots were concentrat ed to the section from 1.5 to 2.5 m (Fig. 1). Dry knots were concentrated at the base (0- 0.5 m) of the logs and at a distance of 1.5-2.5 m (Fig. 2). There were almost twice as many dry knots in the top logs as in the butt logs. V = ni/31(R2 +Rr + (1) Variable Minimum Maximum Average Length (cm) 182 435 304.8 Top diameter with bark (mm) 112 295 159.8 Volume with bark (dm 3 ) 28 221 78 Quality 1 4 2.87 Saw timber -% 13.9 79.4 36.5 V = volume of sawlog 1 = length of sawlog R = radius at butt end r = radius at top end Grade Price, FIM per m 3 a o^nn B 2000 C 1500 Reject 1000 Kärki Predicting the Value of Grey Alder (Alnus incana) Logs Based on External Quality 17 Fig. 1. Fresh knot distribution according to distance from the butt and the type and diameter of a log. Fig. 3. Rotten-knot distribution according to the dis tance from the butt and type and diameter of the log. In large butt logs there were fewer rotten knots than in other types of logs. Rotten knots oc curred most frequently in small-dimensioned top logs and were concentrated in the section from 0.5 to 2.5 m (Fig. 3). Large top logs had more knots than other types Fig. 2. Dry-knot distribution according to the distance from the butt and type and diameter of the log. Fig. 4. Variation in the number of knot types in differ ent types of 3.5-meter logs. of logs did. In large top logs, there were twice as many fresh knots as in small-dimensioned top logs. In general, top logs were more knotted than butt logs were, which was consistent with the process of self-pruning. Rotten knots were more common in top logs than in butt logs (Fig. 4). research articles Silva Fennica 33(1) 18 Table 7. Distance distributions of graded sawn timber from the butt of the logs (all 229 logs). Table 8. Distance distributions of graded sawn timber from the butt (all 229 logs separated into four dimension classes). Average values for different grades show that the grades are not equally distributed throughout the log (Table 7). Class A, i.e. knot-free sawn timber was mainly at the base of the logs. In this respect classes B, C and reject differed from each other only slightly. In Table 8, the logs are divided into four classes. The limit values show that each quality class occurred equally in logs. Overall, it seems that in different types of logs the different grade classes were at the same dis tances. In the butt logs, class C began noticeably farther away from the butt than in top logs. It also seems that the worse the grade class was the greater was also the length of the class. So a rejected piece of sawn wood was usually reject ed totally. The best grade class A was shorter than the other classes, and it could be combined with poorer quality classes (normally with class es B and C). The reasons for grade reductions are shown in Table 9. Number of dry knots was the reason for reduc tion in sawn timber grade in half of the cases. This happens in every quality class. Another very remarkable reduction was seen in quality rejected; discoloration was the reason for quality change in almost half of the cases. Table 9. Reasons for changing the quality in individual piece of sawn wood to a poorer qualityclass (per centage). Table 10. Results of the saw log grading Table 1 1 . Results of the sawn wood grading. One aim of this study was to investigate how suitable the quality classifications of Keinänen and Tahvanainen (Tables 2 and 3) are. In Table 10 the results of the saw log grading are shown. According to the classification (Table 2), only 65 % of the logs could be classified to A, B or C classes and 35 % of the logs belonged to the log grade Rejected (Table 10). The amount of log grade A was modest; only 7.7 % of the logs could be classified to grade A. 35 % of the saw logs were included in log grade B and 21 % to grade C. Grade Average distance from butt (cm) Limit values (cm) A 0-175 0-430 B 62-257 0-435 C 99-295 0-435 Reject 70-271 0-430 Butt logs Top logs Top diameter Grade > 151 mm < 151 mm > 151 mm < 151 mm Average distance (cm) A 0- 91 4-140 0-102 2-127 B 65-252 74-267 46-250 58-253 C 113-299 107-302 95-276 81-304 Reject 78-280 62-251 97-301 58-282 Reason for Class B Class C Rejected Total reduction in grade (A-B) (A,B->C) (A,B,C-Rej.) Number of knots 12.2 17.3 5.6 13.3 Fresh knot 18.1 5.1 - 12.9 Dry knot 52.8 47.4 44.4 50.6 Bark knot 1.9 2.6 2.8 2.1 Rotten knot 7.2 9.0 - 7.2 Cracks 0.3 1.3 - 0.6 Discoloration 7.2 14.1 47.2 12.1 Rot 0.3 3.2 - 1.2 Total 100 100 100 100 Saw log quality Number of logs Volume with bark (m 3 ) Proportion (%) A 21 1.40 7.7 B 66 6.50 35.9 C 49 3.83 21.1 Rejected 93 6.40 35.3 Total 229 18.13 100.0 Sawn wood quality Volume of sawnwood (m 3 ) Proportion (%) A 2.37 35.9 B 2.20 33.3 C 1.30 19.7 Rejected 0.74 11.1 Total 6.61 100.0 Kärki Predicting the Value of Grey Alder (Alnus incana) Logs Based on External Quality 19 Table 12. Correlation coefficients between the main predictors. Fig. 5. Distribution of saw log grades and distribution of sawn wood qualities in various saw log grades. When the results in Table 11 are compared to the log grades measured (Table 10), the main difference is the small amount of Rejected sawn wood. The volume of grade A sawn wood is significantly greater than could be predicted from the log grading. In Figure 5 we can see the relation between grades of the logs and quality of the sawn wood. From the figure can be seen what kind of sawn wood could be obtained from different grades of logs (A. B, C, Rejected). We can also see the average distribution of sawn wood qualities. From the grade-A saw logs we reached over 50 % grade-A sawn wood. The proportion of rejected sawn wood was very small (6 %). The sawn wood quality in log grades B and C do not differ. In both log grades the distribution of sawn wood classes is similar. The log grade Rejected includes the least grade-A sawn wood, but the volume of grades B and Rejected sawn wood are equal to the better sawlog qualities (B and C). Average amount of grade-A sawn wood in the sawlogs was 36 %, for grade-B 33 %, for grade- C2O % and for the Rejected class 11 %. 3.2 Models for Predicting the Value of Logs The correlation matrix between the main predic tors is shown in Table 12. These 9 factors were used when the final models were built. The other factors, such as twist and cracks, were not signif icant, so the correlation matrix for them is not presented. The largest correlations were between FK2 and DKI and between FKI and FK2. These correlations were taken into consideration when the regression models were built. The correla tions between the predictors used in the models are smaller than ±0.252 (between FK3 and TD). The regression models used for predicting the value for grey alder logs based on external qual ity are shown in Table 13. The models were made in order to predict the value of different kinds of logs. There are sepa FK1 FK2 FK3 FK4 DKl DK2 DK3 DK4 TD FK1 1.00 FK2 .522 1.00 FK3 .313 .487 1.00 FK4 .012 .103 .316 1.00 DK1 .035 -.690 -.191 -.090 1.00 DK2 .021 -.004 -.053 -.102 .379 1.00 DK3 .046 -.068 -.150 -.148 .447 .369 1.00 DK4 .102 -.029 -.056 .013 .023 .117 .044 1.00 TD -.209 -.290 -.252 -.119 -.018 -.164 -.038 .054 1.00 FKl = number of fresh knots from base to 0.5 m FK2 = number of fresh knots from 0.5 m to 1.5 m FK3 = number of fresh knots from 1.5 m to 2.5 m FK4 = number of fresh knots from 2.5 m to 3.5 m TD = top diameter of the log DKl = number of dry knots from base to 0.5 m DK2 - number of dry knots from 0.5 m to 1.5 m DK3 = number of dry knots from 1.5 m to 2.5 m DK4 = number of dry knots from 2.5 m to 3.5 m research articles Silva Fennica 33(1) 20 Table 13. Regression models for predicting the value of grey alder logs. rate models for butt and top logs. In explanation of the models there are differences in interpreta tions. In Model 1 there were no significant exter nal signs of quality that could be used in predic tion; obviously, for the knot-free base section of a log the only external signs that can be used are dimensional (e.g. top diameter and length). In Model 2 all the coefficients are negative, which means that all knots (fresh and dry) de crease the value of the stem. This is a direct result of prefering the knot-free timber as the most valuable. Dry knots decrease the value more than fresh knots do. The knottiness at the section from 1.5 m to 2.5 meters is the most significant predictor from quality signs. The standard error is smaller in Model 2 than in Model 1. In Model 3 the coefficient of determination was larger than in Model 2. Top diameter of a log was included in Model 3 and the coefficient of determination improved slightly (37 %). The explanations for the coefficients in this model were logical; the top diameter increases the val ue of a log, and the dry and fresh knots decrease the value. As shown in Model 2, dry knots de crease the value more than fresh knots do. The knotness from the 1.5 m to 2.5 meters was the most significant as in Model 2. Of these six models the standard error for Model 3 was the smallest. Model 4 gives the log value only according to the number of knots. The coefficients of deter mination for different variables are logical in interpretation. Both coefficients are negative, which indicates that knot-free timber is the most valuable. Dry knots influence the log value more than fresh knots do. In Model 5 the top diameter was included, which greatly increased the signif icance of the model. Here the coefficient of de termination was 57 %, and the standard error was also smaller than in Model 4. Residual plots for the two most useful Models (3 and 5) are shown in the figures below (Figure 6). The residual plot for top logs (incl. top diame ter) has the smallest standard error (14.91). Model 5 has a large standard error but the residual is still symmetrical. X-variable B SE B Beta t p > Model 1: Value of butt logs (FIM), R2 = 0.34, SEE = 21.87, n = 136 Intercept -4.20 TD 0.43 0.11 0.34 3.84 0.0002 Model 2: Value of top logs (FIM), R 2 = 0.29, SEE = 15.26, n = 93 Intercept 53.24 DK3 -3.66 1.91 -0.21 -1.92 FK3 -2.19 0.91 -0.26 -2.42 0.0584 0.0177 Model 3: Value of top logs incl. top diameter (FIM), R 2 = 0.37, SEE = 14.91, n Intercept 4.21 DK3 -3.53 1.86 -0.20 -1.90 FK3 -2.16 0.89 -0.26 -2.44 TD 0.36 0.16 0.23 2.23 = 93 0.0615 0.0170 0.0288 Model 4: Value of all logs (FIM), R 2 = 0.31, SEE = 22.29, n = 229 Intercept 67.56 DK2 -7.99 3.43 -0.16 -2.33 FK3 -3.96 0.99 -0.27 -3.99 0.0210 0.0001 Model 5: Value of all logs incl. top diameter (FIM), R 2 = 0.57, SEE = 19.16, n Intercept -7.86 FK3 -1.91 0.88 -0.13 -2.17 TD 0.45 0.05 0.53 8.76 = 229 0.0310 0.0000 B = coefficient of regression, SE B = standard deviation of regression coefficient. Beta = standard coefficient of regression, t = value of t-test, p = level of risk. Kärki Predicting the Value of Grey Alder ( Alnus incana) Logs Based on External Quality 21 Fig. 6. Standardized residuals for Models 3 and 5. 4 Discussion In general, grey alder logs have knots from the base to the top. All types of knots appear and the length of the knot-free section at the base is short. The difference between a dry-knot section and a fresh-knot section is not as obvious as, for example, with birch (Kärkkäinen 1986). In small dimensioned logs there are fewer knots than in larger logs. Especially in large top logs, there are many more fresh knots than in other types of logs. It is interesting that dry knots are concen trated at the base of the logs (0-0.5 m) and in the section from 1.5 to 2.5 m. The number of dry knots in top logs is almost twice that in the butt logs. Rotten knots are concentrated in the sec tion from 0,5 to 2,5 m. Small-dimensioned butt and top logs are less knotty than large logs. If we compare these num bers of knots to the grade classification present ed in Table 2, both averages could be included in grade-B. Large butt logs could be classified as grade-C and large top logs in the rejected cate gory. Nevertheless, this is only a classification according to knottiness, in which the other qual ity factors (bends, cracks, discoloration etc.) are not taken into consideration. Overall, it seems that the different grades are located in compara ble sections along the length in different types of logs. It also seems that the worse the grade class was, the longer was also the length of the class. The most common reasons for decreasing grade were dry knots and discoloration. When the models for predicting the value of grey alder logs are considered, three models seemed to be valid: models predicting the value for butt logs (Model 1), top logs including top diameter (Model 3) and all logs including top diameter (Model 5). Models 2 and 4 (based only on external knot amount) were also logical, but the standard errors were slightly higher because the top diameter was not included. The most important source of error in this material was the subjective sampling of experi mental logs by two loggers. The trees were se lected with the limitation that the logs had to be sawable (stem form). Those trees, from which the logs were cut, obviously were not the aver age trees in the stands as far as quality was concerned. On the other hand, according to the grades in Table 2, they were the best and largest trees. The measurements were made from only two stands. Along with the geographical concen tration of the material, this permits only limited possibilities to apply the results elsewhere in Finland. Nevertheless, the material gives an in dication of the quality of grey alder logs in the Savo-Carelian area. Unfortunately, the size of the knots (beyond classification) was not taken directly into ac count when the logs and timber were classified; if they had been, it would have permitted more detailed analysis of the knot distribution in dif ferent kinds of logs. The knots were measured only partly in order to classify the logs and sawn pieces rightly. The grading used for grey alder logs and sawn timber (Tables 2 and 3) is also one research articles Silva Fennica 33(1) 22 source of error, because it is based on theoretical assumptions and the grade limits are fixed with out empirical studies. This causes error when the grading is used for value predictions. Despite of these error possibilities, the reliability of the re sults can be considered good: one aim of the research was even to investigate the suitability of the classification (Tables 2 and 3) to practise. It is possible to make grading rules for saw logs and sawn wood that are better than those used in these experiments (Keinänen and Tahva nainen 1995). In particular, there are problems in classifying the B and C logs; these types of logs do not differ enough. The problems could be avoided by allowing more dry knots in grades B and C; then the amount of rejected sawlogs would be comparable to that of rejected sawn wood. These classification rules, however, are a step in the right direction for better utilization of alder in Finland. Acknowledgements This study was carried out at Faculty of Forestry of the University of Joensuu. I wish to thank Prof. P. Harstela and Prof. E. Verkasalo for their valuable comments on the manuscript and Dr. J. von Weissenberg for suggestions for revising the English language. References Alestalo, A. & Hentola, Y. 1967. Alder in sulphate pulping. Paper and Timber 50(1): 25-27. (In Finn ish with English summary). Björklund, T. & Ferm, A. 1982. Pienikokoisen koivun ja harmaalepän biomassa ja tekniset ominaisuu det. Folia Forestalia 500. 37 p. (In Finnish) Blomqvist, H. & Nylinder, M. 1988. Relation be tween geometry, yield and quality for sawlogs of spruce. Results from investigations at Rockham mar sawmill. Swedish University of Agricultural Sciences. Dep. of For. Products. Report 202. 51 p. (In Swedish with English summary). Farmer, R.H. 1981. Handbook of hardwoods. Princes Risborough laboratory. London. 243 p. Flinkman, M. 1985. A study of quality defects and their effect on the value of saw timber in grading according to VMR 1-81 and value grading sys tem. Swedish university of Agricultural Sciences. Dep. of For. Products. Report 167.66 p. (In Swed ish with English summary). Grace, L.A. 1993. Evaluating a prototype log sorting system for use in pine sawmills. Swedish Univer sity of Agricultural Sciences. Dep. of For. Prod ucts. Report 237. 43 p. (In Swedish with English summary). Hakkila, P. 1970. Grey alder as pulpwood. Paper and Timber 12: 817-824. Hallock, H. & Galiger L. 1971. Grading hardwood lumber by computer. Res. Pap. FPL 157. USDA Forest Serv., Forest Prod. Lab., Madison, WI. Hanks, L.F. 1976. How to predict lumber-grade yields for graded trees. USDA Forest Serv. Gen. techn. rep. NE-20: 1-9. Harless, T.E.G., Wagner, F.G., Steele, P.H., Taylor, F.W., Yamada. V. & McMillin, C.W. 1991. Methodology for locating defects within hardwood logs and determining their impact on lumber-value yield. Forest Prod. J. 41(4): 25-30. Kärki, T. 1997. Sahauskelpoisen erikoispuun laatu vaatimukset ja käyttö Savo-Karjalan alueella. Fo lia Forestalia 1/1997: (ln Finnish). Kärkkäinen, M. 1986. Malli männyn, kuusen ja koi vun puuaineen oksaisuudesta. Abstract: Model of knottiness of wood material in pine, spruce and birch. Silva Fennica 20(2): 107-116. (In Finnish with English summary). Keinänen, E. & Tahvanainen, V. 1995. Pohjolan jalot puut. Pohjois-Savon erikoispuiden käytön lisää misprojekti. 160 p. (In Finnish). Klinkhachorn, P., Franklin, J.P., McMillin C.W., Con ners, R.W. & Huber H.A. 1988. Automated com puter grading of hardwood lumber. Forest Prod. J. 38(3): 67-69. Lehtonen, 1., Pekkala, O. & Uusvaara, O. 1978. Tech nical properties of black alder (Alnus glutinosa (L.) Gaertn.) and great sallow (Salix caprea L.) wood and pulp. Folia Forestalia 344. 19 p. (In Finnish with English summary). Liljeblad, Ä., Johansson, L.G. & Drake, E. 1988. Qual ity simulation of sawlogs. Träteknik Centrum. Rap port 18812081, Stockholm. (In Swedish with Eng lish summary). Louna, T. & Valkonen, S. 1995. Kotimaisen raaka aineen asema lehtipuiden teollisessa käytössä. Metsäntutkimuslaitoksen tiedonantoja 553. 38 p. (In Finnish) Kärki Predicting the Value of Grey Alder (Alnus incana) Logs Based on External Quality 23 Lövskog: björk, asp och ai i skogsbruk och naturvärd. 1989. Skogsstyrelsen. 255 p. (In Swedish) Lutz, J.F. 1978. Wood veneer: Log selection, cutting and drying. U.S. Forest Service, technical bulletin 1577. 137 p. Mäkinen, T. 1984. Harmaalepän ja tervalepän runkomuoto, kuorimallit ja runkokäyrämallit. Met sänarvioimistieteen pro gradu -tutkielma, Helsingin yliopisto. 60 p. (In Finnish) Meimban, R.J. 1991. Simulation of hardwood saw milling operations. Pennsylvania State Universi ty, Dep. of Forest Resources. Doctoral Thesis. 100 p. Miettinen, L. 1933. Tutkimuksia harmaalepiköiden kasvusta. Metsätieteellisen tutkimuslaitoksen jul kaisuja 18(1): 1-100 p. (In Finnish with German summary) Nakata, K. 1986. Simulation of softwood-log sawing. J. Hokkaido For. Prod. Res. Inst. 3: 15-22 (In Japanese with English summary). Orver, M. 1970. Grading of Scots pine saw timber using objectively measurable factors. Rapp. Inst. Virkeslära Skogshögsk. 66: 1-54. Richards, D.8., Adkins, W.K., Hallock, H. & Bulgrin, E.H. 1979. Simulation of hardwood log sawing. USDA Forest Serv. Res. Pap. FPL 355. For. Prod. Lab., Madison. WI. Schalin, I. 1966. Harmaalepän merkityksestä käytän nön metsätaloudessa. Metsätaloudellinen aikakaus lehti 83(9): 362-366. (In Finnish). Söyrilä, P. 1992. Haapa viilun ja vanerin raaka-ainee na. Paper and Timber 74(8): 622-627. (In Finnish). Uusitalo, J. 1994. Sahatavaran laadun ennustaminen mäntytukkirungoista. Helsingin yliopiston metsä varojen käytön laitoksen julkaisuja 3. 53 p. (In Finnish) Valkonen, S.. Rantala, S. & Sipilä A. 1995. Jalojen lehtipuiden ja tervalepän viljely ja kasvattaminen. Metsäntutkimuslaitoksen tiedonantoja 575.112 p. (In Finnish). Verkasalo, E. 1995. Järeän puun laadun tutkiminen. Seminaariesitelmä metsäteknologian tutkijoiden koulutuspäivillä 1995. Helsinki. 11 p. (In Finnish). Weslien, H. 1983. Value grading of saw timber using objectively measurable factors. Part 1. Grading of pine timber. Sveriges lantbruksuniversitetet. Inst. För virkeslära. Rapport 140: 1-16. Total of 31 references tutkimusartikkeli ^ Metsätieteen itiKciKciuskirid 39 Timo Kärki, K.alle Eerikäinen, Jaakko Heinonen ja KariT Korhonen Harmaalepän (Alnus incana) tilavuustaulukot Kärki, T., Eerikäinen, K., Heinonen, J. & Korhonen, K.T. 1999. Harmaalepän (Alnus incana) tilavuustaulukot. Metsätieteen aikakauskirja I/ 1999: 39—49. Tutkimuksessa esitetään harmaalepän tilavuusmalli yhtälön ja taulukoiden muodossa. Lisäksi tutkimuksessa esitetään kuoren osuuden malli ja dimensiovaatimusten mukaan lasketut tukkiti lavuuden taulukot. Tukkitilavuudet vaihtelevat huomattavasti dimensiovaatimusten mukaan. Eri dimensiokriteerejä vertailtaessa on kuitenkin huomattava, että pienten latvaläpimittaluokkien käyttö sahauksessa edellyttää suoraa runkomuotoa, jotta pienten dimensioiden kannattava sa haus olisi mahdollista. Harmaalepälle lasketut tilavuudet ovat pienempiä kuin vastaavan läpimittaisten koivujen tila vuudet, kun tarkastellaan läpimitaltaan suuria (d > 15 cm) mutta pituudeltaan lyhyitä (h 2o% (2) b x = aiu) b 5% +(1 - ai (x ))b2o%, (3) b x = aim + (1 - Qi(x))btio%, (4) bx = + (1 -Ol(*)) Ö2 , (5) v„ /v= a 0 + a\d (6) Ao = bo + b\x + b\X 2 + h?*3 (7) a\ = + bsx + b(yX 2 + b-jx3 (8) X, % a i ai 1 1,198 10 0,6255 15 0.3686 - 30 0,6702 - 40 0,5861 - 50 0,4016 - 70 0,7920 0,4865 80 0,4463 0,4865 90 0,2533 0,4865 Harmaalepän (Alnus incana) tilavuustaulukot Kärki, Eerikäinen, Heinonen & Korhonen 45 Täten kuorettoman ja kuorellisen tilavuuden suh detta kannon ja suhteellisen korkeuden x välisellä rungon osalla kuvattiin yhtälöllä 9: 3 Tulokset 3.1 Tilavuustaulukot Harmaalepän tilavuustaulukoissa 6, 7, 8 ja 9 esite tyt kuorelliset runkotilavuudet (dm 3 ) on määritetty yhden desimaalin tarkkuudella tilavuudeltaan alle 100 dm3 :n rungoille ja dm3 :n tarkkuudella yli 100 dm3:n rungoille. 3.2 Kuorettoman osuuden mallit Kuorettoman tilavuuden osuuden malliksi estimoi tiin yhtälö: Mallin 10 selitysasteeksi saatiin 0,015 ja läpimitan p-arvoksi 0,035. Mallin selitysaste on alhainen, sil lä kuoren osuus tilavuudesta on lähes vakio puun koosta riippumatta (ks. taulukko 10). Taulukkoon 10 on estimoitu yhtälöllä 10 kuoret toman tilavuuden osuus kuorellisesta tilavuudesta eri läpimittaluokissa. Kuorettoman tilavuuden osuutta mille tahansa korkeudelle estimoivaksi malliksi saatiin yhtälö 11: 4 Tulosten tarkastelu Harmaalepälle ei ole aiemmin esitetty puukohtai sia tilavuustaulukoita, siten tässä tutkimuksessa laa ditun kokonaisrunkotilavuuden taulukon (taulukko 7) vertailukohdaksi valittiin Laasasenahon ja Snell Taulukko 6. Harmaalepän kuorellinen kokonaisrunkotilavuus (dm 3 ) 10 cm.n kannonkorkeudelta määritettynä. v u lv =bo + b\X + b\x 2 + b^x 3 + (£>4 + bsx + b&x 2 + b-jx 3 )d (9) v„/v = 0,9415 - 0,00096 xd (10) v u (x) / v(x) = 0,0489* 3 - 0,1 166 a:2 + 0,0867* + 0,9226 + (0,000 lx3 + 0,0003x2 - 0,0005.x - 0,0008) d (11) h, m 2 3 4 5 6 7 8 9 10 11 12 di.3, cm 13 14 15 16 17 18 19 20 21 22 23 24 25 2 1.4 2,2 3,4 5,1 3 1,4 2,3 3.6 5,4 7,6 10.4 4 1,6 2,5 3,9 5,8 8,2 11,2 14,6 5 1,7 2,7 4,2 6,3 8,9 12,2 15,9 20,2 24,9 6 1,8 2,9 4.6 6.9 9,8 13,3 17,4 22,0 27,2 32,8 38,8 7 2,0 3,2 5,0 7.5 10,7 14,5 19,0 24,1 29,7 35,9 42,4 49,3 56,6 8 3,5 5,5 8,2 11,7 15,9 20,8 26,4 32,6 39,3 46,4 54,0 62,0 70,4 79,1 9 6,0 9.0 12,8 17,4 22,8 28.9 35,6 43,0 50,8 59,1 67,8 77,0 86,5 96,5 107 10 9.8 14.0 19,0 24,9 31,6 38,9 47,0 55,5 64,6 74,2 84,1 94,6 105 117 129 142 11 15,3 20,8 27,2 34.5 42,5 51,3 60,6 70,5 81,0 91,9 103 115 128 141 155 169 185 12 22,6 29.7 37,6 46,4 55,9 66,1 76,9 88,3 100 113 126 139 153 168 184 201 220 240 13 32,3 40,9 50,5 60.9 72,0 83,8 96,1 109 123 137 152 167 183 201 219 239 261 284 14 44,5 54,9 66,2 78,3 91,1 105 119 133 149 165 182 200 218 239 260 284 309 15 59,6 71.9 85,0 98.9 114 129 145 161 179 197 217 237 259 283 308 336 16 77,9 92,2 107 123 140 157 175 194 214 235 257 281 306 334 364 17 99,8 116 133 151 170 190 210 232 254 278 304 332 362 394 18 126 144 164 184 205 227 250 275 301 329 359 391 426 19 156 177 199 221 245 270 297 325 355 387 422 460 20 190 214 239 265 292 320 351 383 418 455 496 Tutkimusartikkelrt Metsätieteen aikakauskirja 1/1999 46 Taulukko 7. Harmaalepän tukkiosan tilavuus (dm 3) 10 cm:n kannonkorkeudelta määritettynä, kun rungon osituskriteerinä käytetään tukin minimipituutta 21 dm ja latvaläpimittaa 13 cm (Keinä nen ja Tahvanainen 1995). Taulukko 8. Harmaalepän tukkiosan tilavuus (dm 3 ) 10 cm:n kannonkorkeudelta määritettynä, kun rungon osituskriteerinä käytetään tukin minimipituutta 30 dm ja latvaläpimittaa 13 cm. männin (1983) esittämä koivun kuorellisen koko naisrunkotilavuuden taulukko. Voidaan todeta, että harmaalepän tilavuusennusteet ovat pienempiä kuin vastaavat koivun tilavuudet, kun tarkastellaan läpi mitaltaan suuria (d > 15 cm) mutta pituudeltaan ly hyitä (h < 10 m) puuluokkia. Tällöin harmaalepän taulukoidut runkotilavuudet suurimmissa läpimitta luokissa ovat enimmillään noin 10 prosenttia pie nemmät suhteessa vastaavien läpimittaluokkien koivun runkotilavuuksiin. Kun tarkastelu kohden netaan pituudeltaan suuriin, mutta läpimitaltaan pie niin taulukoiden puuluokkiin, saadaan nyt esitetyl lä harmaalepän tilavuustaulukolla suurempia run kotilavuuden ennusteita kuin koivun tilavuustaulu kolla. Erotukset ovat enimmillään 11 prosentin luok kaa. Tämä ero pienenee kuitenkin noin kuuteen pro h, m 14 15 16 17 18 (/]3, cm 19 20 21 22 23 24 25 8 44,1 55,5 9 46,7 59,4 71,8 84,1 10 49,4 64,0 77,6 91,1 105 119 11 34,5 51,6 67,9 83,5 98,8 114 129 145 162 12 35,2 54,1 72,3 89,7 107 123 141 158 176 196 216 13 36,1 56,9 77,3 96,5 115 134 152 172 192 213 235 259 14 37,4 60,4 83,1 104 125 145 165 186 208 231 256 282 15 39,1 64,8 89,9 113 136 158 180 202 226 251 278 306 16 41,5 70,4 98,1 124 148 172 196 220 246 273 301 332 17 44,7 77,6 108 135 161 187 213 239 267 296 327 360 18 49,4 86,8 119 149 177 204 232 260 290 321 355 391 19 56,2 98,2 133 164 193 223 252 283 315 349 385 423 20 107 146 180 211 243 274 307 342 378 416 458 h, m 14 15 16 17 18 <*1.3, 19 cm 20 21 22 23 24 25 9 59.4 71,8 84,1 10 64,0 77,6 91,1 105 119 11 67,9 83,5 98,8 114 129 145 162 12 54,1 72,3 89,7 107 123 141 158 176 196 216 13 56,9 77,3 96,5 115 134 152 172 192 213 235 259 14 60,4 83,1 104 125 145 165 186 208 231 256 282 15 64,8 89,9 113 136 158 180 202 226 251 278 306 16 70,4 98,1 124 148 172 196 220 246 273 301 332 17 77,6 108 135 161 187 213 239 267 296 327 360 18 49,4 86,8 119 149 177 204 232 260 290 321 355 391 19 56,2 98,2 133 164 193 223 252 283 315 349 385 423 20 107 146 180 211 243 274 307 342 378 416 458 Harmaalepän (A/nus incana) ti la vu ustau I u kot Kärki, Eerikäinen, Heinonen & Korhonen 47 Taulukko 9. Harmaalepän tukkiosan tilavuus (dm 3 ) 10 cm:n kannonkorkeudelta määritettynä, kun rungon ositus kriteerinä käytetään tukin minimipituutta 20 dm ja latvaläpimittaa 8 cm (Kärki 1997 a). Taulukko 10. Kuorettoman tilavuuden osuus kuorelli sesta tilavuudesta läpimitan funktiona. senttiin, kun puuluokan pituus on 20 m ja rinnan korkeusläpimitta saa arvon 20 cm. Edellä tehtyjen tarkastelujen lähtökohtana ovat olleet tilavuustau lukon äärialueiden runkotilavuudet, joita voidaan pitää ekstrapolointeina. Harmaalepälle laaditun tau lukon keskiosissa (d = 10, h = 10) tilavuudet ovat lähes yhtenevät koivun taulukon vastaavien läpimit ta- ja pituusluokkien tilavuuksien kanssa. Läpimit taluokkaa 20 suurempien luokkien havaintojen vä hyydestä johtuen (ks. taulukko 2) on tälle alueelle tilavuustaulukoissa esitettyihin runkotilavuuden ennusteisiin suhtauduttava varauksella. Taulukoitujen puulajeittaisten runkotilavuuksien erojen syitä voivat olla puulajien toisistaan poikkea va runkomuoto sekä eritoten suurten puiden koh dalla eroa mahdollisesti aiheuttava erilainen kan nonkorkeuden ts. puun kaatokohdan määritys, joka tässä tutkimuksessa oli kaikissa läpimittaluokissa 10 cm maanpinnan tasosta määritettynä. Vertailukoh tana käytetyn koivun tilavuustaulukon (Laasasen aho ja Snellman 1983) tapauksessa tilavuuden in tegraalin aloittamiskohdaksi oli määrätty ylin kaa toa haittaava juurenhaaran niska, joka oli ennustet tu erillisellä yhtälöllä. Kannonkorkeuden minimik si oli asetettu kuitenkin 10 cm. Tässä työssä on esitetty rungon kokonaistilavuu h, m 9 10 11 12 13 14 15 äi3, 16 cm 17 18 19 20 21 22 23 24 25 5 19,6 6 20,7 27,0 33,5 7 14,8 22,0 29,0 36,2 43,6 51,4 8 15,3 23,3 31,2 39,2 47,4 55,9 64,7 73,8 9 15,7 24,7 33,6 42,5 51,6 60,9 70,5 80,5 90,8 102 10 16.1 26,2 36,1 46,0 56,0 66,3 76,9 87,8 99,1 111 123 136 11 16,6 27,9 39,0 49,9 61,0 72,3 83,8 95,8 108 121 134 149 163 179 12 17,2 29,8 42,1 54,2 66,4 78,8 91,4 104 118 132 147 162 178 195 214 234 13 18,1 32,0 45,7 59,0 72,3 85,9 99,7 114 129 144 160 177 194 213 233 255 278 14 19,2 34,6 49,7 64,3 78,8 93,6 109 124 140 157 174 192 211 232 254 277 303 15 37,8 54,2 70,1 85,9 102 118 135 153 171 189 209 230 252 276 301 329 16 59,3 76,5 93,6 111 129 147 166 185 206 227 250 274 299 327 357 17 83,5 102 121 140 160 180 201 223 246 271 297 325 355 387 18 111 131 152 173 195 218 242 267 293 321 351 384 419 19 142 165 188 211 236 262 289 317 347 380 415 453 20 178 203 228 255 283 312 342 375 410 448 489 ä, cm vu/v 1 0,941 2 0,940 3 0,939 4 0,938 5 0,937 6 0,936 7 8 0,935 0,934 9 0,933 10 0,932 11 0,931 12 0,930 13 0,929 14 0,928 15 0,927 16 0,926 17 0,925 0,924 0,923 0,922 0,921 0,920 0,919 0,918 0,917 18 19 20 21 22 23 24 25 Tutkimusartikkelit Metsätieteen aikakauskirja 1/1999 48 den sekä eri dimensiokriteereiden mukaiset teoreet tisen tukkiosan tilavuusestimaatit harmaalepälle (taulukot 6-9). Tukin latvaläpimitan ja pituuden minimivaatimuksilla on huomattava vaikutus teo reettisen tukkiosan tilavuusestimaattiin. Estimoi taessa tilavuuksia Keinäsen ja Tahvanaisen (1995) asettamien minimidimensioiden mukaan tilavuudet muodostuvat huomattavasti pienemmiksi kuin Kär jen (1997 a) asettamilla minimivaatimuksilla. Esi merkiksi 14 cm:n läpimittaisen lepän on oltava 11 m pitkä, jotta se tuottaisi tukkiosaa minimivaatimuk silla 13 cm / 21 dm ja peräti 18 m pitkä minimivaa timuksilla 13 cm /30 dm. Minimivaatimuksilla 8 cm / 20 dm 14 cm:n paksuinen puu tuottaa tukki osaa jo 7 m:n pituisena. Eri dimensiokriteerejä vertailtaessa on kuitenkin huomattava, että pienten latvaläpimittaluokkien käyttö sahauksessa edellyttää suoraa runkomuotoa. Muutoin pienidimensioista puuta on järkevämpi käyttää esimerkiksi pyöreänä puuna rakentamises sa (Boren ym. 1998). Taulukon 10 ja mallin 10 mukaan lepällä kuoren osuus tilavuudesta kasvaa puun läpimitan kasvaes sa. Tulos on päinvastainen kuin esimerkiksi Mäki sen (1984) tutkimuksessa saatu tulos. Ero selitty nee pääosin sillä, että Mäkisen aineistossa suurim mat harmaalepät olivat alle 15 cm läpimitaltaan, kun taas tämän tutkimuksen aineistossa on lähes 30 cm:kin läpimittaisia puita. Kiitokset Kirjoittajat haluavat kiittää prof. Jouko Laasasen ahoa ja VTT Juha Lappia heidän tekemistään lu kuisista merkityksellisistä korjausesityksistä, jotka on otettu huomioon käsikirjoituksessa. Kirjallisuus Alestalo, A. & Hentola, Y. 1967. Leppä sulfaattikeitos sa. Paperi ja Puu 50: 25-27. Baskerville, G.L. 1972. Use of logarithmic regression in the estimation of plant biomass. Canadian Journal of Forestry 2: 49-53. Björklund, T. & Ferm, A. 1982. Pienikokoisen koivun ja harmaalepän biomassa ja tekniset ominaisuudet. Fo lia Forestalia 500. 37 s. Boren, H., Kärki, T. & Lindblad, J. 1998. Pyöröpuutuot teiden markkinat Englannissa ja Saksassa. Joensuun yliopisto, metsätieteellinen tiedekunta, Tiedonantoja 72. 34 s. Bruun, H.H. & Slungaard, S. 1959. Investigation of po rous wood as pulp raw material. 3. Fibre dimensions of several NW European wood species. Paper and Timber 2: 31-34. Crow, T.R. & Schlaegel, B.E. 1988. A guide to using regression equations for estimating tree biomass. Nort hern Journal of applied forestry 5: 15-22. Grosser, D. 1989. Einheimische Nutzhölzer und ihre Verwendungsmoglichkeiten. Institut fur Holzfor schung der Universität Munchen, Munchen. 46 s. Grönros, J., Merra, A. & Mali, J. 1995. Kotimaisten puu lajien ominaisuudet ja saatavuus. Valtion teknillinen tutkimuskeskus. 59 s. Hakkila, P. 1970. Basic density, bark percentage and dry matter content of grey alder (Alnus incana). Commu nicationes Instituti Forestalls Fenniae 71(5). 33 s. Ilvessalo, Y. 1947. Pystypuiden kuutioimistaulukot. Com municationes Instituti Forestalls Fenniae 34(4). 149 s. Kalela, E.K. 1936. Tutkimuksia Itä-Suomen kuusi-har maaleppä -sekametsiköiden kehityksestä. Suomalaisen Kirjallisuuden Seuran kirjapaino Oy, Helsinki. 179 s. Keinänen, E. & Tahvanainen, V. 1995. Pohjolan jalot puut. Pohjois-Savon erikoispuiden käytön lisäämispro jekti. 160 s. Kärki, T. 1997 a. Sahauskelpoisen erikoispuun laatuvaa timukset ja käyttö Savo-Karjalan alueella. Metsätie teen aikakauskirja - Folia Forestalia 1/1997: 37^18. 1997 b. Haapa- ja leppätukkien kysyntä, hankinta ja laatu. Joensuun yliopisto, metsätieteellinen tiedekun ta, Tiedonantoja 53. 78 s. —, Maltamo, M. & Eerikäinen, K. 1998. Static models for describing tree stock and quality of grey alder fo rests in eastern Finland. Käsikiijoitus. Laasasenaho, J. 1982. Taper curve and volume functions for pine, spruce and birch. Seloste: Männyn, kuusen ja koivun runkokäyrä- ja tilavuusyhtälöt. Communi cationes Instituti Forestalls Fenniae 108. 74 s. & Snellman, C-G. 1983. Männyn, kuusen ja koivun tilavuustaulukot. Metsäntutkimuslaitoksen tiedonan toja 113. 91 s. Lahtinen, A. & Laasasenaho, J. 1979. On the construc tions of taper curves by using spline functions. Selos te: Runkokäyrän muodostaminen splini-funktiolla. Communicationes Instituti Forestalls Fenniae 97(8). 63 s. Harmaalepän (Alnus incana) tilavuustaulukot Kärki, Eerikäinen, Heinonen & Korhonen 49 Lappi, J. 1993. Metsäbiometrian menetelmiä. Silva Ca relica 24. 182 s. Louna, T. & Valkonen, S. 1995. Kotimaisen raaka-ai neen asema lehtipuiden teollisessa käytössä. Metsän tutkimuslaitoksen tiedonantoja 553. 38 s. Miettinen, L. 1933. Tutkimuksia harmaalepiköiden kas vusta. Metsätieteellisen tutkimuslaitoksen julkaisuja 18(1). 100 s. Mäkinen, T. 1984. Harmaalepän ja tervalepän runkomuo to, kuorimallitjarunkokäyrämallit. Metsänarvioimis tieteen pro gradu -tutkielma. Helsingin yliopisto. 60 s. + liitteet. Nousiainen, J., Puranen, J. & Tiihonen, P. 1973. Koivut ukkipuiden kuutioimis-menetelmä. Communicationes Instituti Forestalls Fenniae 79. 53 s. Routala, O. & Sihtola, H. 1934. Tutkimuksia lepän käyt tömahdollisuuksista selluloosan raaka-aineena. Acta Chemica Fennica 7: 113-119. Saarsalmi, A. 1995. Nutrition of deciduous tree species grown in short rotation stands. University of Joensuu, Faculty of Forestry, Research Notes 37. 60 s. & Mälkönen, E. 1989. Harmaalepikön biomassan tuo tos ja ravinteiden käyttö. Folia Forestalia 728. 16 s. —, Palmgren, K. & Levula, T. 1983. Viljelylepikon alku kehitys ja biomassaan sitoutuneiden ravinteiden mää rä. Metsäntutkimuslaitoksen tiedonantoja 107. 33 s. —, Palmgren, K. & Levula, T. 1985. Leppäviljelmän biomassan tuotos seka ravinteiden ja veden käyttö. Folia Forestalia 628. 24 s. —, Palmgren, K. & Levula, T. 1991. Harmaalepän veso jen biomassan tuotos ja ravinteiden käyttö. Folia Fo restalia 768. 25 s. —, Palmgren, K. & Levula, T. 1992. Harmaalepän ja rauduskoivun biomassan tuotos ja ravinteiden käyttö energiapuuviljelmällä. Folia Forestalia 797. 29 s. SAS. 1992. SAS/STAT software: changes and enhance ments, release 6.07. SAS Institute Inc., Cary, N.C., Technical report P-229. 620 s. Schalin, I. 1966. Harmaalepän merkityksestä käytännön metsätaloudessa. Metsätaloudellinen aikakauslehti 83(9): 362-366 s. Varmola, M. & Vuokila, E. 1986. Pienten mäntyjen tila vuusyhtälöt ja -taulukot. Folia Forestalia 652. 24 s. Vuokila, Y. 1960. Lehtikuusen kuutioimisyhtälöt ja -tau lukot. Communicationes Instituti Forestalis Fenniae 51(10). 89 s. 34 viitettä The photos in front cover: Top left: 27-year-old grey alder forest in North Karelia, G= 32 m 2 /ha, Hdom =l6 m. Photo:Timo Kärki Middle left: Western red alder (Alnus rubra) panels of B-quality. Photo: Inland Wood Specialties, L.P. United States Down left: Furniture made of red alder. Photo: Pollmeier GmbH. Germany Top right: Planing of grey alder wood. Mäntyharjun Viilu Oy. Photo:Timo Kärki Middle right: 3-layer plate made of grey alder. At left A-quality and at right B-quality. Photo:Timo Kärki Down right: Furniture made of Finnish grey alder. Photo: Oy Esse-Möbel Ab, Ähtävä. METSÄNTUTKIMUSLAITOKSEN TIEDONANTOJA 764, 2000 FINNISH FOREST RESEARCH INSTITUTE, RESEARCH PAPERS 764,2000 ISBN 951-40-1722-6, ISSN 0358-4283