UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Single and integrated use of forest lands in British Columbia - theory and practice Sahajananthan, Sivaguru 1995

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1995-060586.pdf [ 5.86MB ]
Metadata
JSON: 831-1.0075321.json
JSON-LD: 831-1.0075321-ld.json
RDF/XML (Pretty): 831-1.0075321-rdf.xml
RDF/JSON: 831-1.0075321-rdf.json
Turtle: 831-1.0075321-turtle.txt
N-Triples: 831-1.0075321-rdf-ntriples.txt
Original Record: 831-1.0075321-source.json
Full Text
831-1.0075321-fulltext.txt
Citation
831-1.0075321.ris

Full Text

SINGLE AND INTEGRATED USE OF FOREST LANDS IN BRITISH COLUMBIA THEORY AND PRACTICE -  by SIVAGURU SAHAJANANTHAN B.Sc. Special (Hons.). University of Colombo, Sri Lanka 1971 Diploma in Forestry. Indian Forest College, India 1976 M.Sc. University of Oxford, U.K. 1981  A THESIS SUBMITTED N PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Forest Resources Management) We accept this thesis as conforming to the required standard  THE UNIVERSITY OF BRITISH COLUMBIA October 1995  © Sivaguru Sahajananthan, 1995  In presenting this thesis in  partial fulfilment of the requirements for an advanced  degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department  or  by  his  or  her  representatives.  It  is  understood  that  copying  or  publication of this thesis for financial gain shall not be allowed without my written permission.  (Signature)  Department  of  The University of British Columbia Vancouver, Canada  Date  DE-6 (2188)  ‘‘  !  I7Ya  11  ABSTRACT This study deals with the multiple use management of forests.  The main  objectives of the study are i) to review the literature on economic theory of multiple use and examine various approaches taken by foresters to practice multiple use, and ii) to compare, with respect to timber supply, rent and selected environmental indicators, two alternative forest land use systems under three timber management intensities (basic, medium and high). A review of the literature suggests that benefits accruing from multiple use forestry can be measured in terms of rent and the provision of amenity values (non-timber goods and services). There is an inverse relationship between timber rent and the flow of natural amenity values. Current forest practices in British Columbia (BC) attempt to maintain a constant flow of natural amenity values by retaining certain structural elements in the landscape through a system of resource emphasis rules (RER). These stringent RERs may lead to high amenity flow and low rent. These theoretical findings were empirically tested by simulation with spatially explicit models, ATLAS and SIMFOR, in a sub-unit of the Revelstoke Timber Supply Area, Revelstoke 1 in British Columbia. The study shows that the opportunity costs of integrated management, as currently practiced, is equivalent to 60% of potential sustainable timber supply. Analysis of RERs shows that universally applied adjacency constraints reduce sustainable timber harvests and rents by 58% and 65% respectively, when compared to an unconstrained base case.  Visual quality constraints reduce  sustainable timber harvests to as low as 9 % of the base case. The study also estimates that, in the absence of RERs, 46% of the net operable area of Revelstoke 1 can produce evenflow volumes equivalent to those currently produced by the whole area of the Unit. This area can be further reduced to 35% with intensive timber management.  111  Two types of multiple use systems, integrated use (IU) system and single use (SU) system were devised. The IU system treats the whole operable’ area as an integral unit, while the SU system has a timber zone and an integrated zone. This research shows that the SU system rent is higher than that of IU system at all assumed management intensities, and at “high intensity”, can be as much as 216% relative to that of the IU system at basic intensity. Rents from both systems were found be to very sensitive to discount rates. Rent from the IU system is found to be more sensitive to changes in prices than from the SU system but their relative performance does not change. Environmental indicators suggest that the IU system leads to higher fragmentation of critical “interior forest” wildlife habitats and consequent loss of amenity values compared to the SU system. The road density in the SU system is found to vary from 65% to 68% of that of the IU system under “basic” and “high” intensities. As for the protection of biodiversity, the SU system is likely to help maintain a system of small reserves scattered throughout the working forest as a complement to the system of large reserves in the protected areas. This research has implications for short timber supply, rent and forest stewardship. In the short term, the SU system will release for immediate harvest, highly productive sites carrying high timber volumes hitherto locked up by adjacency constraints.  Timber production zones will facilitate the creation of secure tenure  arrangements designed to protect public investments and provide incentives for the private sector to invest in timber production on public land.  The study, also,  demonstrates the high potential for single use management through land use zoning as a strategy to balance economic and environmental values from British Columbia’s forests and offers an innovative method for achieving sustainablility of timber and non-timber resources.  iv  TABLE OF CONTENTS Abstract  Table ofcontents  .  ii  iv  List oftables  xii  List offigures  xiv  Acknowledgment  xx  Dedication  xxi  1. INTRODUCTION  1  1.1 Background  1  1.2 Research problem  4  1.3 Organization of the thesis  7  2 ECONOMIC THEORY OF MULTIPLE USE  12  2.1 Introduction  12  2.1.1 Firm theory aspects of multiple use  12  2.1.2 Capital theory aspects of multiple use  12  2.1.2.1 Appropriate discount rates  13  V  2.2 Cost function in multiple use production  16  2.2.1 Interdependence in multiple use production  16  2.3 Case for specialization in production  19  2.3.1 Site productivity  21  2.3.2 Diseconomies ofjointness and diseconomies of scale  23  2.3.3 Management efforts  25  2.3.4 Empirical evidence  26  2.4 Marginal costs in multiple use production  26  2.5 Aggregate human welfare  29  2.5.1 Timber rent  30  2.5.2 Timber rent and amenity values from a multiple use forest  31  2.6 The economic problem facing the forest manager  33  3 MULTIPLE USE MANAGEMENT IN PRACTICE  39  3.1 Multiple use forest harvest models  39  3.1.1 Single stand models  40  3.1.2 Forest level harvest models  41  3.1.2.1 Strata based models  42  vi 3.1.2.2 Area based models  .42  3.1.3 ATLAS (A Tactical Land Analysis System) model  43  3.2 Landscape pattern modeling  44  4 EMPIRICAL STUDY  46  4.1 Introduction  46  4.2 Description of the study area  46  4.2.1 General  46  4.2.1.1 Access units  47  4.2.1.2 Stand groups  47  4.2.1.3 Site quality  47  4.2.2 Resource use conflicts  48  4.2.3 Current management practices  49  4.2.3.1 Resource emphasis areas  51  5 RESEARCH METHODOLOGY  53  5.1. Introduction  53  5.2 Total resource planning  53  5.2.1 Temporal modification of forests  54  vii 5.3 Methodology for stand level analysis  55  5.3.1 Silvicultural treatments  55  5.3.1.1 Artificial regeneration  55  5.3.1.2 Pre-commercial thinning  56  5.3.1.3 Commercial thinning  56  5.3.1.4 Pruning  57  5.3.1.5 Fertilization  57  5.3.2 Silvicultural regimes  57  5.3.2.1 Simulation modeling for growth and yield  59  5.3.2.2 Simulation modeling for bucking and sawing  60  5.3.3 Selection of rotation age  61  5.3.4 Stand level economic analysis  62  5.3.4.1 Assumptions used in the stand level economic analysis  63  5.4. Methodology for forest level analysis  65  5.4.1 Alternative land use systems  65  5.4.2 Simulation modeling for forest level harvesting  65  5.4.2.1 Planning horizon  66  viii 5.4.2.2 Parameters determined by ATLAS simulation modeling  66  5.4.2.3 Harvest scenarios  67  5.4.2.3.1 Scenario modeling current management practices  67  5.4.2.3.2 Scenario modeling alternative land use systems  67  5.4.2.3.3 Scenario modeling enhancement of non-timber values within the timber zone  68  5.4.3 Simulation modeling for landscape pattern responses  69  5.4.4 Forest level economic analysis  70  5.4.4.1 Assumptions used in the forest level analysis  71  5.4.4.2 Price sensitivity analysis  73  5.5 Indicators of net benefit to society  73  5.5.1 Economic parameters  74  5.5.2 Environmental parameters  74  6 STAND LEVEL ANALYSIS  76  6.1 Introduction  76  6.2 Analysis of growth and yield  76  6.2.1 Growth patterns of stand groups  76  6.2.2 Effect of silvicultural treatment regimes on age of maximum mean annual increment  77  ix 6.2.3 Determination of rotation ages  .77  6.2.4 Effect of silvicultural treatment regimes on volume and diameter at breast height  80  6.3 Stand level economic analysis  83  6.4 Implications for forest level analysis  88  7 FOREST LEVEL ANALYSIS  90  7.1 Current management practices  90  7.1.1 Impact of resource emphasis rules on timber supply  90  7.1.1.1 Timber emphasis  93  7.1.1.2 Wildlife emphasis  94  7.1.1.3 Visual quality  94  7.2 Timber supply under alternative land use systems  95  7.2.1 Alternative land use systems  95  7.2.1.1 Integrated use system  95  7.2.1.2 Single use system  96  7.2.2 Timber supply under alternative land use systems  96  7.3 Intensive timber management on a 240 year planning horizon  99  7.3.1 Economic parameters  99  x 7.3.1.1 Maximum evenflow volume  .99  7.3.1.2 Timberrent  101  7.3.1.3 Sensitivity of rent to discount rates  102  7.3.1.4 Sensitivity of rent to changes in price of logs  103  7.3.1.5 Delivered wood costs  104  7.3.2 Environmental parameters  107  7.3.2.1 Seral stages  107  7.3.2.2 Ecosystems represented in seral stages  110  7.3.2.3 Edge habitats  112  7.3.2.4 Influence of old-growth edge on regeneration  116  7.3.2.5 Patch sizes  119  7.3.2.6 Harvest pattern in old-growth  125  7.3.2.7 Density of roads  126  7.4 Enhancement of non-timber values in the timber zone  128  7.4.1 Impact on even-flow volume  128  7.4.2 Impact on rent  130  8 DISCUSSION  131  xi 8.1 Opportunity cost of resource emphasis areas  131  8.2 Implications for timber supply  134  8.2.1 Even-flow volumes  134  8.2.2 Rent  135  8.3 Implications for wildlife  138  8.4 Implications for visual quality  142  8.5 Implications for forest stewardship  143  8.6 Implications for Forest Renewal Plan  146  8.7 Implications for short term timber supply  148  8.8 Strategy for the future  150  9 SUMMARY AN]) CONCLUSIONS  152  BIBLIOGRAPHY  165  APPENDIX  174  xl’  List of Tables 8  Table 1 List and description of codes used in the tables and figures Table 2 Site indices of selected species  @ 50 years  48  Table 3 Resource Emphasis Rules for Revelstoke TSA  50  Table 4 Distribution of resource emphasis areas in Reveistoke 1 and the whole TSA .51 .  Table 5 Summary of resource emphasis rules and their area of application in Revelstoke 1 Table 6 Silvicultural treatments examined for development of silvicuitural regimes  52 59  Table 7 Harvest scenarios in the SU and the IU systems with intensive timber management  68  Table 8 Harvest scenarios for the timber zone with enhanced non-timber values  68  Table 9 Seral stages distinguished in the land use systems and their description  70  Table 10 Silvicultural treatment regimes selected for stand level economic analysis  84  Table 11 Additional silvicultural treatment regimes selected for Douglas-fir  85  Table 12 Silvicultural treatment regimes showing economic feasibility  87  Table 13 Additional silvicultural regimes for Douglas-fir showing economic feasibility  88  Table 14 Ecosystem types represented in very old-growth as percent area of total land base at the start and end of the 240 year planning horizon  110  xlii  Table 15 Average area of edge habitat as percent area of old-growth habitats, and average area of regeneration edge as percent of regeneration area over a 240 year planning horizon  115  Table 16 Average area covered by the four types of patches in the very old-growth seral stage at the end of 240 year planning horizon  120  xiv  List of Figures Figure 1  Types of production possibilities for two products on a tract of land  Figure 2  Relative productivities of sites determining the selection of production of goods and services  Figure 3  20  21  Optimal production of two products on two sites with varying site productivities  23  Figure 4  Optimal production with two substitute products  25  Figure 5  Marginal cost of timber production with no constraints compared separately with production under visual quality and wildlife constraints  29  Figure 6  Rent under the IU and the SU systems  31  Figure 7  Relationship between timber rent and amenity in multiple use systems  33  Figure 8  Growth curves for Douglas-fir (SI=19), cedar (SI=21) and spruce (SI=18)  78  Figure 9  Effect of silvicultural treatments on growth curves of Douglas-fir (SI=19)  78  Figure 10 Effect of silvicultural treatment regimes on age of maximum MAI of Douglas- fir (SI=19) Figure 11 Rotation ages of stand groups used in the regenerated forest  79 79  Figure 12 Effect of silvicultural treatment regimes on Volume and DBH of final harvest in Douglas-fir on good and medium sites (SI=19)  80  Figure 13 Effect of silvicultural treatment regimes on Volume and DBH of final harvest in Douglas-fir on poor sites (S112)  81  xv Figure 14 Effect of silvicultural treatment regimes on Volume and DBH of final harvest in redcedar on good and medium sites (SI=2 1)  81  Figure 15 Effect of silvicultural treatment regimes on volumes and DBH of final 82  harvest in redcedar on poor sites (SI=1 3) Figure 16 Effect of silvicultural treatment regimes on volume and DBH of fmal harvest in spruce on good and medium sites (SI=1 8)  82  Figure 17 Effect of silvicultural treatment regimes on volumes and DBH of final 83  harvest in spruce on poor sites (SI=10) Figure 18 Percent of clear lumber in pruned logs of different diameter classes in Douglas-fir (SI=19) that is commercially thinned @70 years and  85  harvested @130 years Figure 19 Percent increase in value of pruned logs of different diameter classes in Douglas-fir (SI=19) that is commercially thinned  @ 70 years and  harvested @130 years  86  Figure 20 Redcedar (SI=21): Actual and affordable costs for selected silvicultural treatment regimes  86  Figure 21 Redcedar (SI=21): Feasibility as indicated by discounted net revenues of selected silvicultural treatment regimes  87  Figure 22 Impact of resource emphasis rules on timber supply  91  Figure 23 Impact of resource emphasis rules on rent  92  Figure 24 Opportunity cost in terms of timber values of selected resource emphasis rules when applied individually to the entire land base  92  xvi Figure 25 Impact of adjacency, disturbance and cover constraints of timber emphasis 93  rule on timber supply Figure 26 Integrated use system showing composition of harvest area by forest types  98  over the 120 year planning horizon Figure 27 Single use system showing composition of harvest area by forest types over the 120 year planning horizon  99  Figure 28 Impact of intensive timber management with integrated and single use systems on maximum evenflow volume on a 240 year planning horizon  100  Figure 29 Impact of intensive timber management with integrated use and single use systems on rent (2 % discount rate on a 240 year planning horizon)  101  Figure 30 Impact of intensive timber management with integrated use and single use systems on rent (0 % discount rate, 240 year planning horizon)  102  Figure 31 Impact of intensive timber management with integrated use and single use systems on rent (4% discount rate, 240 year planning horizon) Figure 32 Percent change in rent to increase and decrease in log prices by 12%  103 104  Figure 33 Delivered wood costs in integrated use and single use systems with medium intensity management showing its components viz., hauling cost, harvest system cost, and road construction and maintenance costs  105  Figure 34 The IU and the SU systems at medium management intensity showing periodic cost of road construction and maintenance on a 240 year planning horizon  106  xvii Figure 35 Single Use showing the distribution of seral stages for basic intensity management over a 240 year planning horizon  109  Figure 36 Integrated Use showing the distribution of seral stages for basic intensity management over a 240 year planning horizon  109  Figure 37 Single Use showing the distribution of ecosystem types within very oldgrowth seral stage (>240 years) for basic intensity management on a 240 year planning horizon  111  Figure 38 Integrated Use showing the distribution of ecosystem types within very oldgrowth seral stage (>240 years) for basic intensity management on a 240 year planning horizon  112  Figure 39 The SU and the IU systems showing area of edge habitat (as percent area of old-growth (>120 years) and very old-growth (>240 years)) at basic intensity management on a 240 year planning horizon  113  Figure 40 The SU and the IU systems showing the area of edge habitat (as percent area of old-growth (>120 years) and very old-growth (>240 years) for medium intensity management over a 240 year planning horizon  113  Figure 41 The SU and the IU systems showing the distribution of edge habitats (as percent area of old-growth (>120 years) and very old-growth (240 years) for high intensity management over a 240 year planning horizon  114  Figure 42 The SU and the IU systems showing the area of regeneration affected (as percent of the regeneration area) by old-growth and very old-growth edges for basic intensity management on 240 year planning horizon  118  xviii Figure 43 The SU and the IU systems showing the area of regeneration affected (as percent of the regeneration area) by old-growth and very old-growth edges for medium intensity management on 240 year planning horizon  118  Figure 44 The SU and the IU systems showing the area of regeneration affected (as percent of the regeneration area) by old-growth and very old-growth edges for high intensity management on 240 year planning horizon  119  Figure 45 The SU system showing the distribution of patch sizes in very old-growth (>240 years) for basic intensity management over a 240 year planning horizon  122  Figure 46 The IU system showing the distribution of patch sizes in very old-growth (>240 years) for basic intensity management over a 240 year planning horizon  123  Figure 47 The SU system showing the distribution of patch sizes in very old-growth (>240 years) for medium intensity management over a 240 year planning horizon  123  Figure 48 The JU system showing the distribution of patch sizes in very old-growth (>240 years) for medium intensity management over a 240 year planning horizon  124  Figure 49 The SU system showing the distribution of patch sizes in very old-growth (>240 years) for high intensity management over a 240 year planning horizon  124  xix Figure 50 The IU system showing the distribution of patch sizes in very old-growth (>240 years) for high intensity management over a 240 year planning horizon  125  Figure 51 Percent of old-growth (OG) retained with integrated use and single use systems at the end of the 240 year planning horizon  126  Figure 52 Road density with integrated use and single use systems at basic, medium and high management intensities showing average length of roads maintained per period, and constructed during the planning horizon  127  Figure 53 Impact of wildlife and visual quality emphases on volume from the timber zone  129  Figure 54 Impact of wildlife and visual quality emphases on rent from the timber zone  130  xx  ACKNOWLEDGMENT First of all, I would like to sincerely thank my supervisor Professor David Haley for selecting me as his graduate student, for employing me as his teaching and research assistant, for suggesting an excellent research topic for my PhD and for giving me continued guidance throughout my stay as a graduate student at the University of British Colunthia. I am very much indebted to him. Jam very grateful to my supervisory committee members Professors John Nelson, Philip Burton and Hans Schreier. I am much obliged to Professor John Nelson who, as a member of my supervisory research committee and as interim supervisor for one year provided the digitized data on Revelstoke 1, trained me to use ATLAS, provided me financial support at dfflcult times, and gave me continued guidance at every stage of my research project. The help given by Professors Burton and Schreier was also invaluable. They, in spite of their busy schedules have always found time to do a thorough and critical review of my draft reports. Their constructive criticism, followed byfruiful discussion have helped a lot in fine tuning my thesis. I appreciate the support and encouragement given for my research effort by several other members of the faculty, notably Dean Clark Binkley and Professors Peter Marshall, Peter Pearse, Gordon Weetman, G.C. Van Kooten and Les Lavkulich. I am particularly grateful to Professor Peter Marshallfor employing me as his teaching assistant. Jam thankful to computer wizards: Tim Shannon and Dave Daust -for helping me with ATLAS and SIMFOR software. They were always there to help me whenever I had any problems. I am grateful to many people from outside the University of British Columbia, particularly fromB.C. Ministry of Forests and Ministry of Environment and Parks andfrom the private industrial sector, who have helped me in many ways to conduct this research. Foremost among them is Mr. David Raven, District Forest Officer for the Reveistoke District. In spite of his busy schedule, he was always available to lend supportfor my research. Another key person is my friend Ken Polsson who was responsible for running the TASS model and providing me with growth and yield data. Addtionally, Mr. Jim Blake and Mr. Ken Talbot of the B.C. Ministiy ofForests in Revelstoke, Mr. Barry Wagnerfrom Downie Timber LTD., and Mr. Cohn Pike from Bell Pole & Co. deserve special recognition and thanks. In fact, every one I have come in contact with in the B.C. Ministry of Forest offices at Reveistoke and in Victoria have been extremely helpful. I also would like to thank Professor ilan Vertinsky for providing me a place to study at the FEPA research unit. My thanks are also due to Ms Mabel Yee, the secretary at FEPA, and to our former graduate student secretary Ms Natalie Cole and the present one Ms Lily Liew who have always taken the interests ofthe students to heart. I appreciate the support given by my friends and graduate student colleagues at FEPA and at Forest Operations. I also would like to thank my wfe ‘s maternal uncle Ponnampalam Balasundaram who supported and encouraged me in many ways to help me achieve my goal. My thanks are also due to the Forest Department and the Government of Sri Lankafor releasing me on no-pay leave. Lasz, but not the least, I would like to thank my dear wfe Thulasifor her moral andfinancial support. She has endured lots of hardship, and sacrficed many things for the sake of my studies and to make my 23 year old dream coming true. If not for her I may not have survived my return to graduate studies after many years ofprofessional career.  xxi  DEDICATION This research work is dedicated to my belovedparents Sivaguru and Nallammah and to my beloved wife Thulasi who sacrificed a lot oftheir happinessfor the sake ofmy education.  1. INTRODUCTION  1.1 BACKGROUND  Forests produce a multitude of goods and services.  However, historically, in  British Columbia (BC), except for the protection of some areas set aside as parks and ecological reserves, forests have been managed mainly for the production of timber. A general philosophy prevailed that when a forest is managed for timber, other resource values will follow in sufficient amounts. In other words, management for other resource values was passive rather than active. Multiple use management became mandatory in BC with the passage of the Ministry ofForests Act which for the first time spelled out the objectives of the Ministry (Ministry of Forests Act, 1978. RS Chap. 272, Section 5). While changes in practice were slow to follow, the rise of the environmental movement in the 1980’s turned concern for “the environment” into a mainstream phenomenon which has since been reflected in political action. The capability of a tract of land to produce goods and services which meet human demands depends on its biological and physical characteristics. The production of any mix of products is in delicate balance.  Attempts to produce more of one product  invariably affect others either positively or negatively. It is not possible to satisfy all societal demands simultaneously from the same tract of land. Land management for multiple products involves many trade-offs. The task of the land manager is to seek an optimum strategy which provides the highest attainable level of social welfare, or satisfaction, for the resources (land, labor and capital) available.  2 Governments and the bureaucracy have responded to increasing demands on forest resources by imposing additional constraints on timber harvesting, and by completely withdrawing an increasing number of areas from timber production. This has resulted in land use problems of crisis proportions for the forest industry of BC, which is a major driving force of the province’s economy. Many of the steps taken since 1990 were precipitated by the work of the British Columbia Forest Resources Commission (Peel 1991). The Provincial Government that came to power in 1991 introduced a number of new administrative procedures, regulations and statutes designed to address land use problems in the forestry sector. These include: •  Commissioner on Resources and Environment (CORE)  •  CORE Land Use Charter  •  Protected Area Strategy  •  Forest Practices Code  •  Forest Renewal Plan. CORE established the “process objectives” to ensure sustainability of natural  ecosystems and the economy they support through meaningful public participation and due consideration of the concerns of the aboriginal people (CORE 1992. Chap. 34, BC Reg. 234/92).  The CORE Land Use Charter sets out the fundamental principles of  environmental, social and economic sustainability that should guide natural resource planning and management in British Columbia (Owen 1994). This charter was adopted in principle by the BC Government in June 1993. The objectives of the Protected Area  3 Strategy are (i) to protect viable representative samples of natural diversity of the province, and (ii) to protect the special natural, cultural heritage and recreational features of the province (Government of B.C. 1993). It has set a target to increase the amount of protected area in the province from the current 8% to 12 % in the form of Parks, Ecological Reserves and Wilderness Areas, by the year 2000 (Owen’1994). The Forest Practices Code sets out regulations and standards that will ensure good stewardship for the management of forests for multiple uses (Forest Practices Code of British Columbia Act, SBC Chap 41 Vol 2 Bill 40, 1994).  The Forest Renewal Plan  provides for  investment in long term growth and diversity of the forest and for sustaining forest dependent communities (British Columbia Forest Renewal Act, SBC Chap 3 Vol 1 Bill 32, 1994). Based on the above policies, the Government has released the CORE’s strategic regional plans for four regions: Vancouver Island; Cariboo-Chilcotin; West Kootenay Boundary; and East Kootenay. These regional plans attempt to solve land use problems by designating zones (protected, special management, integrated and dedicated) for management with varying intensities of timber management under a multiple use frame work as specified in the Forest Practices Code. On one end of the scale are the protected areas with absolutely no industrial resource extraction or timber production which would compromise natural ecosystem functions. On the other end are the dedicated zones where management interventions which may compromise natural ecosysterñ functions to meet human values are allowed but basic environmental quality is maintained (that is, intensive management of timber production is allowed or, indeed, promoted).  4  Current regulations on multiple use emphasize the simultaneous production of multiple goods and services from the same tract of land. The intensity of production for particular uses varies according to the management emphasis for the resource in question. Management involves trade-offs among the various uses in order to achieve an “optimum” mix. Since the physical relationships between the inputs and outputs for even single uses are not well known, not to mention complex interactions between uses, this is not an easy task.  It is further confounded by the fact that many of the products forests  produce do not have values established in the market place. The designation of land use categories under the CORE Land Use Plans are based on a broad regional scale. No attempt has been made to analyze in detail land use on the scale of a management unit. The purpose of this research is to explore alternative approaches to land use planning on a discrete forest management unit. Specifically, the intention is to compare the integrated multiple use approach, in which each hectare of land is managed for several uses simultaneously, to the zoned multiple use approach incorporating a mosaic of single or specialized uses across the forest.  1.2 RESEARCH PROBLEM  There are two basic approaches to multiple use management: i) the management of all hectares of land within a unit for an optimum mix of products; and ii) the zoning of land within a unit for specialized uses (single or dominant use) and managing it as a mosaic or composite of single, specialized uses.  5 Current management practices in BC essentially belong to the first category where an attempt is made to produce a mix of products simultaneously. The proportion of the products in the mix are being constantly challenged by various interest groups.  In  response to the pressures from these groups, attempts are being made by the government to resort to zoning of areas with varying intensities of management for identified resource values within a frame work of multiple use. However, neither the current practices nor the proposed zoning system under CORE Regional plans advocate the establishment of timber zones exclusively for the production of timber. Multiple use theory suggests that optimum use of a forest will generally include some areas where a single use dominates and other areas where several uses will prevail (Gregory 1987). However, practical applications of multiple use theory on a forest level are rare and, in my opinion, are deficient in several respects: i) They do not consider multiple use production in a spatial and temporal manner. New perspectives, particularly those relating to biodiversity, require more attention to spatial and temporal details. ii) There is inadequate attention to the production of non-market goods such as wildlife and visual quality. iii) Too little consideration is given to the impact of management intensification on optimum multiple use planning. iv) Little attention has been paid to the “external” benefits of dominant use zoning, particularly for timber.  For example, dedicating areas for timber production may  6 release additional areas for management of other resource values and thus mitigate the incidence and severity of costly land use conflicts. There is an urgent need to take into account the above factors and investigate the practical economic implications of multiple use practices in spatial and temporal contexts. Most of the guidelines that are aimed at either maintaining or enhancing other resource values from the forest have spatial and temporal implications which directly affect the timber supply. Sustainability of natural ecosystems and the economy they support are so interdependent  that one cannot be achieved without the other.  Caring only for  ecosystems without any concern for the economy can only compromise its very survival (Brundtland 1987). The objective of this research is to investigate the economic implications (as measured by timber supply and rent) of regulations relating to the production of timber and two non- timber resource values (wildlife and visual quality), under two alternative land use systems. The study quantifies inputs to forest management and investigates some of the spatial and temporal changes in selected indicators of landscape pattern (environmental indicators) and discusses their environmental significance. Finally, this study also sets up an analytical framework which will help guide decision making when considering future options. The following alternative land use systems will be investigated on a discrete management unit.  7 •  An integrated multiple use system in which all products are produced simultaneously on every hectare ofland.  •  An integrated multi:ple use system with intensive managementfor timber.  •  A multiple use system with timber zonesfor the production oftimber.  •  A multzle use system with intensive managementfor timber in timber zones.  1.3 ORGANIZATION OF THE THESIS The thesis is organized into nine chapters. The first gives a historical background to the problem and states the specific objectives of the research. The second chapter is a review of literature relating to multiple use theory. In the third chapter, the development of planning models (including area based planning) for multiple use management are discussed. The fourth gives a description of the area selected for empirical study. The fifth discusses the analytical methodology, assumptions and data collection. The sixth chapter addresses the stand level analysis of the problem, while the seventh chapter reports and discusses the results of the forest level analysis. The eighth chapter is a discussion of the results and speculates on implications and zoning policy for British Columbia. The last chapter gives a summary and the conclusions that were reached from this research. A list and description of codes used in the tables and figures in this study are given in Table 1. A glossary of terms used in the thesis is given in Appendix.  8  Table 1 List and description of codes used in the tables and figures Code  Description  1  basic timber management intensity  2  medium timber management intensity  3  high timber management intensity  A  natural amenity values under the JUsystem natural amenity values under the SUsystem  b  basic  c13  western redcedar (Sfr43)  c21  western redcedar (SI=21)  const/240PH  constructed during 240 year planning horizon  CT  commercial thinning  CT(112)  commercial thinning with removal of 1/2 volume ofstand  CT(1/3)  commercial thinning with removal of1/3 volume ofstand  Cw  western red cedar  DBH  diameter at breast height  Df  Douglas-fir  EAF  equivalent annualflow  ESSF  Engelmann Spruce Subalpine Fir Zone  Fl  first application offertilizer  fl2  Douglas-fir (S1 12)  f19  Douglas-fir (5fr49)  F2  second application offertilizer  g&m  good and medium sites  h  high  Haul_C  hauling cost  Hsys_C  harvest system cost  -  9 Table 1 (Continued). See title Code  Descrivtion  IU  integrated use system  1  low  ICHmw  Moist Warm Interior Cedar Hemlock Zone  ICHwk  Wet Cool Interior Cedar Hemlock Zone  m  medium  mature  mature seral stage (61 120 years)  MAI_max  age ofmaximum MAI  MC  marginal cost oftimber production under the lUsystem  MC  marginal cost oftimber production under the SUsystem  MCT  marginal cost oftimber production with no constraints  MCVQ  marginal cost oftimber production with visual quality constraints  MC  marginal cost oftimber production with wildlife constraints  Mgt  management  Natural Regen  naturally regenerated crop after harvesting  notreat  no silvicultural treatment  OG  old-growth  OG_harvested  old-growth harvested during planning horizon  OG_remaining  old-growth remaining at end ofplanning horizon  old  old-growth seral stage (121 240 years)  open/pd  opened during a period  OppCost  opportunity cost  p  poor sites  P1  pruning withfirst lfi (3m height)  P2  pruning with second lfi’ (Sm height)  pat_i  habitat patch size 1 (0 -100 ha)  pat_2  habitatpatch size 2 (101 500 ha)  -  -  -  -  -  10 Table 1 (Continued). See title Code  Description  pat_3  habitatpatch 3 (501 1000 ha)  pat_4  habitatpatch 4 (> 1000 ha)  PC12  precommercially thinned to 1200 sph  PC5  precommercially thinned to 500 sph  PC8  precommercially thinned to 800 sph  PCT  precommercial thinning  PH  planning horizon  P1  artjficially planted  pole  pole seral stage (21 50 years)  r  regenerated crop after harvesting  R  rent under the lUsystem  -  -  rent under the SU system REA  resource emphasis area  regen  regeneration seral stage (0 20 years)  Regimes  silvicultural regimes  RER  resource emphasis rule  Revi  Reveistoke 1 or Akolkolex drainage  Road_C  cost ofopening and maintaining roads  RSV  reserves orforest lands where timber is not harvested  slO  spruce (S1=10)  s18  spruce (SI=18)  Sec-growth  second-growth  SI @50  site index at 50 years  sph  stems per hectare  spp  species type  -  11 Table 1 (Continued). See title Code  Description  SU  single use system  Se  Engelmann spruce  Timber  timber emphasis equivalent to wildlife basic quality RER  Tvolume  total volume  U  Unconstrained timber harvest  veryold  very old-growth seral stage (>240 years)  VM  visual quality modfIcation RER or basic level visual quality  vol  volume  VPR  visual quality partial retention RER or high level visual quality  WC1  wildlfe class 1 equivalent to wildlife high quality RER  WC2  wild4fe class 2 equivalent to wild4fe low quality RER  WC3  wildlfe class 3 equivalent to wildlife medium quality RER  WLD  wildiands or economically operable forest lands  Wilfe  wildflfe  12  2 ECONOMIC THEORY OF MULTIPLE USE  2.1 INTRODUCTION The economics of multiple use forest management deals with four fundamental questions: i) Is production of more than one good technically feasible? ii) If feasible, is it socially and economically desirable to produce it? iii) What are the best combination of products? iv) What are the economically efficient levels of production for each combination of products that will generate the highest social value?  These questions have been  extensively researched (Gregory 1955; Hagenstein and Dowdie 1962; Hartman 1976; Pearse 1969; Walter 1977; Bowes and Krutilla 1985, 1989; Swallow et al. 1990; Vincent and Binkley 1993). The most extensive assessment of multiple use forestry can be found in the book by Bowes and Krutilla (1989). Essentially, the economic theory of multiple use is based on the theory of the multiproduct firm and on capital theory.  2.1.1 Firm theory aspects of multiple use Multiple use theory is based on the theory of the multiproduct firm because it is concerned with combining quantities of factor inputs to produce outputs of goods and services and their pricing and output decisions. Production of goods and services from a forest is commonly related to the conditions of the timber growing stock and on the fixed input, land.  2.1.2 Capital theory aspects of multiple use Multiple use theory is based on capital theory as it deals with valuation of costs and benefits over time. Capital stock in forestry is the timber growing stock. The flow of goods  13  and services from a multiple use forest at any point in time is strongly dependent on the growing stock conditions at that time. The output mix is often manipulated by adjusting the level of growing stock through timber harvests. Comparison of the impact of various management options on a multiple use forest would require summing up of costs and benefits spread over time to a single value. This is achieved by using appropriate rates of interest which represent the opportunity costs of capital over time. Future values can be obtained by compounding the present values with an appropriate rate of interest. In a similar manner, present value can be obtained by discounting the future values with the appropriate rate of interest. Discounting is used to mean any process of revaluing a future event, condition, service or product to give a present equivalent (Price 1993).  2.1.2.1 Appropriate discount rates The issue of choosing of the appropriate social discount rate may play a critical role in intertemporal decisions concerning the use of environmental resources.  Two  concepts i) the social opportunity cost (SOC) of capital, and ii) the social rate of time preference (SRTP) shape the discount rate. The rate based on SOC of capital measures the value to society of the next best alternative investment in which funds might otherwise  have been employed. The SRTP  is defined as a rate that reflects the  community’s marginal weight on consumption at different points in time (Kula 1992). In an ideally functioning market, the interest rate equals both the marginal rates of time preference and return on capital. In practice, however, market failures and government policies lead to divergence between these rates (Munasinghe 1993).  The economic  14 theory suggests that in the choice of a social rate of discount, the two rates, i.e. the SOC and the SRTP, should play a joint role (Feldstein 1964; Marglin 1963). Higher discount rates may discriminate against future generations as projects with social costs occurring in the long term and the net social benefits occurring in the near term, will be favored by high discount rates (Munasinghe 1993). This is likely to lead to the rejection of projects with delayed returns under high discount rates thus discriminating against future generations. During the last two decades, there has been a lot of discussion about the ways to deal with time, particularly when it involves intergenerational comparisons. Suggestions include: i) the use of very low discount rates (as low as 0 %, Mishan 1976; or 2 %, Hampson 1972; Hartman 1990), ii) the use of declining discount rates (Cline 1992), and iii) a modified discounting system which treats all generations as equal (Kula 1992). As an alternative to the various methods of discounting, the imposing of sustainability constraints have been proposed by some economists (Jacobs 1993; Munasinghe 1993; Pezzy 1992). The aim of these constraints is to ensure that the overall stock of capital is preserved or enhanced for future generations.  There are several arguments for and  against these suggestions. Most of the empirical studies done to estimate the social discount rate are based on short periods (Lowenstein 1987). Studies on appropriate discount rates to long time horizons are rare. One particular study of relevance is that of Cropper and Portney (1991) who conducted a survey on the value of lives saved at various points over a long time horizon. They found that the discount rate dropped from 7% to 3.5% for 50 years and to  15 0% for 100 years. A phenomenon that shows some similarity to forest management, as far as delayed returns and environmental implications are concerned, is global warming. Management activities that prevent global warming in the future are supposed to show benefits only after about 250 to 300 years. For economic analysis of global warming Cline (1992) recommends the use of 1.5% as the discounting rate. This is based on several theoretical and empirical studies. For similar reasons, Thompson et al. (1992) used a discount rate of 1.5% to evaluate alternative strategies for evaluating the rehabilitation of the backlog of unstocked forest lands in British Columbia. The B.C. Ministry of Forests uses a discount rate of 4% for its analysis (Stone 1993). Most of these analyses generally involve only a few decades, though sometimes are extended to a rotation of about 120 years (Stone 1994; Laing and McCulloch 1993). For government investment projects, the Congressional Budget Office (CBO) in the United States uses a discount rate of 2% and employ a sensitivity analysis, showing the results for ± 2 percentage points around “the” rate, i.e., 0% to 4% (Hartman 1990). In this research, forest level analysis was done for a 240 year planning horizon with a real data set having old-growth and second-growth stands which, upon harvest, are converted to modified forest. The effect of intensive management can only be seen over the second rotation starting after 120 years. At a discount rate of 4% (used by the MOF), a single dollar return after 240 years is worth only 0.008 cents. As such, at a 4% discount rate the analysis cannot discriminate between benefits of incremental silviculture.  In  order to capture these benefits this research used a discount rate of 2% and provides a sensitivity analysis at 0% and 4%.  16 2.2 COST FUNCTION IN MULTIPLE USE PRODUCTION The production of goods and services from a forest can be technically defined by a cost function which represents the least cost of producing any mix of outputs. It can be represented as, C(Q:R,M) where  Q is the mix of outputs, R is a vector of variable input prices and M is  a vector of fixed inputs. The cost function is useful in computing some basic measures such as marginal costs, separable costs and isocost curves. Marginal cost is the incremental cost of increasing production (by one unit) while keeping the levels of production of the other products constant. Separable cost refers to the increase in cost of including an additional output (new output) in an overall production mix. Isocost curves represent the production possibilities achievable with least cost means; they are particularly useful in evaluating a set of cost efficient possibilities for a given budget expenditure or for developing equal cost management options for various budget levels.  2.2.1 Interdependence in multiple use production Multiple use forestry is characterized by jointness in production of marketable and non-marketable products (goods and services). The versatility of the factors of production allow the transformation of one product into another along a production possibility or transformation curve.  The shape of the transformation curve at any point in time is  dependent on the degree of technical interdependence between the products.  The  interdependence could be for a specific level of outputs (a local measure) or for all levels of  17 output (a global measure). interaction (that is,  Multiple use products, based on their local measure of  specific to certain levels of output) can be classified as mutually  exclusive, complements, substitutes or independents. Examples of these production possibility curves are illustrated in Figure 1 (a through 1). i) Mutually exclusive uses: In this case the uses are entirely incompatible and it is not possible to produce these products simultaneously from the same tract of land. An example of a production possibility curve for timber production and wilderness areas is illustrated in Figure 1(a). ii) Complementary uses: Complements show a positively sloping curve where one form of production enhances the other. That is, an increase in production of one leads to a reduction in the marginal cost of producing the other. For instance, timber production improves some wildlife habitats and vice versa at certain levels of output or timber roads provide access for public recreation. This is illustrated in Figure 1(b). iii) Substitute uses:  Substitutes show a negatively sloping curve where one form of  production competes with the production of the other and therefore replaces it in various ways. In this case, an increase in production of one leads to an increase in marginal cost of producing the other. The degree of substitutability is indicated by the rate of transformation. A concave (to the origin) production possibility curve with increasing marginal rate of transformation of one product for the other indicates competitive uses. The curvature reflects the degree of competitiveness and varies over the range of possibilities. This is the most common relationship seen in multiple use forests. Figure 1(c) illustrates the production possibility curve of two competitive uses, timber  18 production and recreation. A convex (to the origin) production possibility curve implies a decreasing marginal rate of transformation of one output for another indicating highly conflicting uses.  That is, successive increments in the output of one product can be  accommodated with progressively smaller sacrifices of the other. Figure 1(d) illustrates a convex production possibility curve for timber production and aesthetic values which are considered to be highly conflicting. In this case, as we successively increase the production of one product the sacrifice in terms of the other product become smaller. A negatively sloping straight line indicates constantly substitutable uses, where the marginal rate of transformation is constant. In this case, the marginal cost remains constant.  Figure 1(e) illustrates the production possibility curve for constantly  substitutable uses of a forest for sawlogs and pulpwood. iv) Independent uses: In this case the production of one product does not have any effect on the production of others. This situation is illustrated by two curves that are at right angles to each other. The marginal cost of producing one product is independent of the rate of production of the other products. Figure 1(f) illustrates the production possibility curves for two independent products, watershed protection and recreation. In a multiple use forest production system, there will always be eventual diminishing returns to scale caused by the limited land area.  In addition, the type and  degree ofjointness and differing scales of production are associated with varying levels of economies and diseconomies. For instance, global complements result in economies in production while global substitutes cause diseconomies of production. The economies of scale, a local measure specific to a particular output mix, may change with output mix. In  19 multiple use, timber production may be associated with scale economies to some output level but it is soon affected by diminishing returns to scale caused by the other products in the mix. The cost of production is also affected by the differences in site productivity. A site may be more productive for a particular product or a set of products and, therefore, its costs of production will be less than that for less productive sites.  2.3 CASE FOR SPECIALIZATION IN PRODUCTION  Some of the peculiarities in costs exhibited by multiple use forests suggest that it is more economically efficient to specialize in the production of some goods and service (either as single use or as dominant use) rather than producing all products from every hectare of land (integrated use). The peculiarities are: i) differences in site productivity, ii) diseconomies of joint production and economies of scale, and iii) responsiveness of products to management efforts.  These are discussed in the following sections.  illustrations are taken from Bowes and Krutilla (1989).  The  20  Timber /(m 3 halyr)  Some  Wilderness area (ha)  1(b) Complementary uses  1(a) Mutually exclusive uses  Timber /halyr) 3 (m  Timber lhalyr) 3 (m  Grazing  Aesthetic values /ha  /ha  1(c) Competing uses  1(d) Highly conflicting uses  Watershed quality  Timber /halyr) 3 (m  N  Recreation days  Pulpwood (m3/halyr)  1(e) Constantly substitutable uses  1(f) Independent uses  Figure 1 Types of production possibilities for two products on a tract of land. (Source. Pearse (1990)  21  2.3.1 Site productivity A site is considered to be more productive than another when more of a certain product can be produced at lower cost. The decision to allocate the production of specific products to their respective sites is generally based on the relative productivity of those , 2 1 and q sites. This situation is illustrated in Figure 2. In the production of two products q Site A is more productive in q while site B is more productive in q . In this case it is better 1  1 q  B  A  2 q Figure 2 Relative productivities of sites determining the selection of production of goods and services.  to produce both products separately under single use management, allocating site A for the production of q and site B for the production of q . For example, some sites with low 1  timber productivity may be good for some types of recreation (for example, skiing) and vice versa. Sometimes one site can be more productive with respect to the production of two products than another, but vary in the relative productivities of each product. An example is 2 than site B, is 1 arid q illustrated in Figure 3, where site A, though more productive in q  22  . 1 relatively more productive for q than for q  Suppose it is decided to produce the equal  quantities of q 1 and q 2 from both sites A and B, then the point of production can be indicated by EA and EB, on the isocost curves A and B, respectively. At the point of ), the ratio of the 2 1 and q optimal production mix (that is, at least cost of producing q marginal cost of producing q 1 and q at site A will be equal to the ratio of the marginal costs 2 of producing q 2 at site B (that is, MC_Aqi/ MC_Aq 1 and q  ). From the 2 MC_Bqi / MC_Bq  point EB, production can move along the isocost curve of site B to the point XB by giving up one unit of q 1 in proportion of the slope of the curve. 2 and increasing the quantity of q Similarly, on the isocost curve of site A, there can be movement in production from EA to 1 in proportion to the slope XA by gaining one unit of q by giving up q  of the isocost curve.  The slope of the isocost curve of site A is flatter than that of the curve B. Therefore, the gain on q 1 on site A for exchanging equal units of 1 in site B is much more than the loss in q • Thus it can be seen that without losing any q 2 we can produce more of q 1 at the same q a q and 2 a 1 from site A and cost. Therefore, it will be economically efficient to produce produce  b 1 q  and q from site B. This shows that specialization will be more economical in  the production of some goods.  23  1 q  A b 1 q  a  2 q  2 q  Figure 3 Optimal production of two products on two sites with varying site productivities.  2.3.2 Diseconomies of jointness and diseconomies of scale The diseconomies ofjointness exhibited by substitute products with convex isocost cost curves favor specialization even if there are no differences in site productivity. An 3 for two ,C 1 , and C 2 example is illustrated in Figure 4. In this figure three isocost curves C identical sites for the production of two products q 1 and q 2 are shown. The required overall production mix is q 1 and q which is labeled as Y. The balanced product mix where each site produces q /2 and ci 1 2 is indicated by the point labeled X which is on the highest . It can be seen that 3 isocost curve C . Therefore, the overall cost of producing Y will be 2C 3 3 at X 1 and ci are equal) that meets C the tangent (when the marginal cost of production of q 1 at XA. By producing at points XA 2 at XB and the isocost curve C meets the isocost curve C  24 and XB the same quantity of q 1 and q can be produced at a lower cost. This can be inferred C2 + 3 is greater than 1 C This indicates that . from the position of the isocost curves since 2C a given quantity of the products q 1 and q 2 can be produced at a lower cost when the production of both goods are specialized (single use in site A for q , and dominant use of 2 site A for q ). 2 In the case of constantly substitutable products, if the isovalue curve has the same slope as that of the isocost curve any combination of products will give the same value. In these situations there will be indifference between specialized and integrated production. But when the isovalue curve does not have the same slope as the isocost curve, corner solutions are possible indicating that specialization will be more beneficial than integrated production. Specialization, however, may not always lead to least cost production. Specialization requires an increase in the scale of production which may result in diseconomies of scale. For example, specialized production of timber may be more profitable with very large cut blocks. But these large cutblocks may. substantially affect other ecosystem functions such as biodiversity and hydrological regimes. Thus, specialization in the production of substitute products is desirable only when the diseconomies ofjointness outweigh the diseconomies of scale (Bowes and Krutilla 1989).  25  1 q  1 q  Figure 4 Optimal production with two substitute products  2.3.3 Management efforts Any expenditures on the factors of production aimed at increasing the level of production of a single product, or a set of products, from a multiple use forest can be termed a management effort. Any management effort applied to increase production of an output from a planning unit will invariably affect the production of others. However, the degree of responsiveness of different products may vary. Vincent and Binkley (1993) have shown that provided that there are no rapid diminishing returns, even with two sites of identical  26 productivity, optimum management will tend towards specialization in production of the product that responds most to management effort.  2.3.4 Empirical evidence The peculiarities in cost of production of a multiple use forest discussed in sections 2.3.1 through 2.3.3 show that any fixed amount of products .such as timber and visual aesthetics (or wildlife) that are competitive may be produced at a lower cost when produced separately as a single use or as a dominant use than it can by producing it in an integrated manner.  Empirical evidence supporting the establishment of zones that  specialize in one or a small number of goods and services are rare. However, research in the Olympic peninsula by Sedjo and Bowes (1990) on the different ecological regimes of New Forestry, such as green tree retention and set asides, have shown that set aside regimes give a higher return than other options. This is also a form of specialization where an attempt is made to set aside a parcel of land for non-timber uses.  2.4 MARGINAL COSTS IN MULTIPLE USE PRODUCTION In a multiple use forest, active production of one product such as timber (say a quantity q ) may positively or negatively affect the production of amenity services or other 1 2 non-timber goods (this relationship can be represented by a function q  =  ). Low p(q ) 1  levels of timber production may enhance the production of some of the non-timber values because of its complementary nature at these production levels.  At high levels of  production some of the non-timber products may be negatively impacted. Maintaining a high level of timber production while maintaining a high level of amenity services would  27 increase the cost of producing timber. Thus, conceptually, the cost of timber production in multiple use forests can be assumed to be made up of two components: , without regard to the amenity 1 i) the cost of producing timber by the least cost means, kq services , where k is constant and q 1 is the quantity of timber produced; and ) that results from 1 ii) the cost of maintaining amenity services above the base level p(q single purpose management. The cost function for timber and wildlife, for example, can be represented thus: 1 ,q 1 C(q ) = kq 2  +  2 c(q  -  1 refers to the cost 2 p(q where: c(q )) refers to the cost of producing wildlife, and kq 1 -  of timber production Marginal cost also can be considered to be made up of two components as follows: 1 dc/dq  =  k- c’i’  where: k refers to the cost of timber production, c’p’ reflects the change in the cost of maintaining wildlife services at the chosen level, and measures the expenses of compensating for changes in base level of wildlife services caused by increased  timber production. At low levels of timber production, the marginal cost may be even less than k when ) is positive and p’ is also positive. At higher levels of timber production 1 i(q  ji’  will be  negative and therefore marginal cost will be higher than k. Wildlife and visual quality constraints increase the marginal cost of timber production. At low levels of timber production wildlife constraints may cause a negative marginal cost for timber production due to the complementary nature of wildlife and timber  28  production.  However, at high levels of timber production small increases in timber  production may increase the marginal cost of timber production several fold. Visual quality constraints, due to their highly conflicting nature with timber production, may increase the marginal cost of timber production from the moment the timber harvesting begins. The marginal cost of producing timber with visual quality constraints rises much faster than that with wildlife constraints. The marginal costs of timber production with no constraints, with visual quality constraints, and with wildlife constraints are illustrated in Figure 5. In this example, in order to illustrate the impact of wildlife and visual quality constraints, the marginal cost of timber production (with no constraints) is assumed to be constant, though due to diminishing marginal returns it is likely to show an initial decrease and then an increase with increasing timber production. In Figure 5, the curves MCT, MCw, and MCVQ represent marginal costs of timber production with no constraints, with wildlife constraints, and with visual quality constraints, respectively. At low levels of timber production, the MC may be even lower than the marginal cost of producing timber as a single use (MCT) due its complementary nature. But at higher levels of timber production, it increases rapidly. The MCVQ, on the other hand rises rapidly with the commencement of timber production and is steeper than MC. The figure illustrates that the cost of timber production with visual quality and wildlife habitats may be very expensive when compared to single use timber production.  29  Cost ($/m)  3 C  2 C MCT  Cl  1  Figure 5 Marginal cost of timber production with no constraints compared separately with production under visual quality and wildlife constraints. (Refer to Table] (Page 8) for description ofcodes)  2.5 AGGREGATE HUMAN WELFARE Rent is the residual value of land or the net return the land can generate by combining all factors of production. Multiple use management will be efficient when all variable factors of production are combined with the fixed factor of land to generate the maximum rent. If all products and services produced by the forest can be priced, then it can be said that the net rent generated is a measure of human welfare. Unfortunately, since many of the products produced by forests are non-marketable, rent alone, as measured by market prices, is an unsatisfactory measure of welfare. Forest products from multiple use  30 can be broadly categorized into timber and amenity. Amenity refers to naturalness and to the goods and services that are associated with naturalness. A major component of the forest rent is the rent generated from timber production. Therefore, timber rent along with amenity values generated by a tract of forest land could be considered to be a good indicator of aggregate human welfare.  Maximizing the sum of these two  values is likely to  maximize the net benefit from the forest.  2.5.1 Timber rent The amount of timber rent generated depends on the relative values of timber produced and the costs of production. If timber values are assumed to be constant, then the timber rent will be directly proportional to the cost of production. The marginal cost of timber production under unconstrained single use (MC ) and under integrated use (MC) 5 is illustrated in Figure 6. The marginal costs rise with increases in timber production due to diminishing returns to scale, but rise more sharply under integrated use. In this example it is assumed that the price of timber is fixed and determined exogenously. At a fixed price , the rent generated will be equal to the area above the marginal cost curve and below the 1 P price curve. This will be equal to the area A + B for integrated use (IU), and to the area A + B+C+D  +  E for single use (SU). If evenflow volume harvesting is practiced then the rent  from the SU system will only be greater than that of the IU by the area C  +  D. Removal of  the evenflow constraint, or changing to area based management, will help maximize rent by an addition of the area E to rent.  31  Price  ) 3 ($/m 1 MC  1 P  /  1 T  Figure 6 Rent under the IU and the SU systems. (Refer to Table 1 (page 8)for descrztion ofcodes).  2.5.2 Timber rent and amenity values from a multiple use forest In a multiple use forest there is a general relationship between the production of timber and amenity. Amenity is not much affected at low levels of timber production, but affected severely at high levels of timber production.  This is illustrated in Figure 7.  Amenity value is shown on the left vertical axis. In the case of integrated use, amenity is represented by harvesting constraints on every hectare of timber production, while for single use it is represented by equivalent areas being withdrawn from timber production.  32 Timber rent for the total area is shown on the right vertical axis. At low levels of timber production, the impact of timber production on amenity values is so low there is hardly any difference between single use and integrated use. In fact in this case the single use is more like integrated use where timber production is limited to very small areas. At 100% timber harvest, there are no amenity constraints and again there is no difference between integrated use and single use as all areas are single use. Thus at this level of timber production, the rent generated under both forms of use will be the same. The difference in rent between the single and integrated use widens when there is a need for high degree of amenity value and high volume of timber production. The figure 7 shows the rent generated for single use (Rsu) and integrated use (R) for varying levels of timber production. Rent generated by single use at high levels of timber production with high amenity constraints will be very much higher than that for integrated use. At any level of production, while the amenity values are maintained constant, the rent from single use can be increased up to its intensive margin by varying the intensities of management. At high levels of timber production, the dispersed harvests in integrated use may even negatively affect the level of amenity values compared to single use. Thus the curve A 1 is shown at somewhat lower levels. In multiple use management, the flow of timber at any point in time is fixed by an annual allowable cut (AAC) after taking into account a fixed flow of amenity services. Timber rent generated by a fixed flow of timber and a fixed flow of amenity from a multiple use forest is likely to vary based on the economics of production. For example, if the timber flow is fixed at level  Q on the X axis (Figure 7), rent generated under single use will  2 generated under integrated use. This is 1 which will be higher than the rent R be equal to R  33  because single use production takes into account factors such as productivity of the site, diseconomies of jointness and management effort. Though amenity levels are expected to be maintained at constant levels through resource emphasis rules set at single stand levels under both single and integrated uses, there is always the possibility .of diseconomies of scale in timber production under integrated use that may reduce the level of amenity from . 2 1 to A A  Natural amenity values I Total area  1 A 2 A  Q 0%  /total area) 3 Timber harvest (m  100%  Figure 7 Relationship between timber rent and amenity in multiple use systems. (Refer to Table 1 (page 8) for description ofcodes).  2.6 THE ECONOMIC PROBLEM FACING THE FOREST MANAGER Human values put on many of the goods and services that are produced by forests appear to have increased dramatically over the last decade with increasing environmental  34 awareness. With the prevailing set of values it can be said that the public demands the production of some products (timber, recreation, hunting etc.) in amounts above those that can be produced under natural conditions, while demanding that the production of other products such biodiversity, wilderness, water yield and quality, air quality etc. be maintained at natural levels. Production of all goods and services in nature are in a state of equilibrium that can be likened to the economist’s Pareto efficient condition where: i) it is not possible to increase the production of any one of the outputs without decreasing the output of another, and ii) where it is impossible to achieve the same level of a specific mix of output with less of any one input without correspondingly increasing the level of another input. Thus, the production of any good or service beyond its natural level will affect the production of some other goods or services.  The question of the technical limit of  disturbance will be determined by the resiliency of the ecosystem. The general economic problem facing the forest manager is, therefore, to fmd the optimum mix of outputs that provides the greatest overall value of net benefits while maintaining the integrity of the ecosystem. A forest ecosystem has integrity if its structure and species composition, the rate of its ecological processes and its ability to resist change in the face of disturbance or stress are within the characteristic range exhibited historically by that ecosystem (Kimmins 1994).  Bowes and Krutilla (1989) define the economic  problem for the forester as ‘the selection of a sequence of harvests and stocks that will maximize the net present value from all current and future flows of harvests and services from the area’.  For integrated use management this applies to every hectare of land,  35  whereas for single use management it applies to a patchwork of single uses within a larger unit of land. The foresters problem can be represented as follows: Max {B(Q) C(Q)} -  selecting that output mix Q* that will generate the highest net value. Where:  Q = output mix, B(Q) = benefit from output mix Q, and C(Q) = cost  of producing output mix Q. A product mix is optimal when it satisfies the condition B(Q*) B(Q) >= C(Q*) C(Q) -  -  Economically efficient multiple use management should take into consideration all the costs and benefits associated with jointness in production, site productivity, economies of scale, and management effort. This requires the following information: i) the assessment of production possibilities for various combinations of uses; ii) the relative demands for and economic values of the various goods and services produced by the forest; and iii) relative response of outputs to management effort. Unfortunately, for a practicing forester, the inherent uncertainty and the long time horizon associated with forest production make it al.most impossible to have perfect information in all these areas. The problem is further complicated by the scale of planning, as to whether it is local, regional or national. Foresters have circumvented this problem by attempting to maximize timber harvests while maintaining certain structural features of the growing stock that are thought likely to ensure a continuous flow of some of the non-timber  36 resource values through time. These structural features are maintaind through time by designing harvesting patterns with temporal and spatial constraints. Some of the structural features maintained in the forest are to: i) maintain a certain percent of various seral stages at all points in time (forest cover constraints); ii) limit spatial disturbance at any point in time by limiting size of cut blocks (or opening size) and by having adjacency constraints (or exclusion periods); and br iii) limit temporal disturbance by having a minimum harvest age. When such modifications are done, it is assumed that there will be a steady flow of non-timber benefits from the forest. Thus, the practical economic problem for foresters managing for multiple use centers on maximizing the net present value from the flow of timber harvests subject to certain resource constraints. Timber production with integrated use (IU), can be represented as follows:  37  =Max>[A+6  ‘13,I-L--o  Since the flow of non timber values A from the whole forest is constant over time, the maximization problem can be expressed as, -  wa  i (h, ‘C a [o ‘Hö 1  where, = net present value from the multiple use forest t = time of harvest = identity of the harvest cut block at time t n number of cutblocks discount factor =1 / (1 + r)t where r equals the rate of discount A = net present value of non timber goods and services. This is constant over time =  = =  H  =  price of timber per cubic meter at time t volume harvested from harvest cutbiock i at time t  volume of timber harvested at time t  H,=h,, aj = constraints to timber production in harvest cutblock i. It is constant over time but not space c, = cost of producing timber at time t from cutbiock i with constraints a  Timber production under single use or specialized production can also be represented as follows: =  [A+o  ‘13,H-o’ 1=1  Sire the flow ofinflimber values A is constant over tin the imximization problem lSecome  =  i[ö  ‘RIi-oEC,(h,, 1=1  Ct = cost ofxkxing given volun oftinixr in harvest cutbiock i with no constraints.  38 The difference between timber production under integrated use and single use is that there are no constraints on the harvest of timber in the area selected for single use timber management, therefore, no symbol “a” in the equation.  It is assumed that the value of  amenity “A” that is provided by constraints “a” in integrated use is provided by a nontimber production zone in single use. In this research no attempt is made to maximize net present value from timber production under any of these constraints.  But an attempt is made to represent the  constraints in a spatially explicit manner in two types of multiple use production (Integrated Use and Single Use) scenarios and compare their feasible solutions with a view to see whether they behave in the theoretically predictable way.  39  3 MULTIPLE USE MANAGEMENT IN PRACTICE Since the 1 940s the ways of practicing multiple use management have been a subject of much controversy. Dana (1943) and Pearson (1944) argued for different ways of practicing multiple use forestry. Dana argued for the production of all possible goods and services from the same parcel of land at the same point in time, whereas Pearson argued for segregation of small parcels of land for special functions, and wanted to consider multiple use on a larger unit of land. Burton (1995) proposes a combined system of zoned single and integrated use designed to maximize net benefit from a discrete management unit. In North America, the dominant use option was very much prevalent until the 1960’s. Latterly, due to a new environmental awareness, there has been a trend towards Dana’s approach which favors integrated management. This philosophy has been further strengthened by the concept of “New Forestry”, which advocates the modification of forest practices in an attempt to achieve old growth like structure in all forested areas (O’Keefe 1990). In British Columbia, current multiple use management practices reflect these trends and emphasize integrated management in which an attempt is made to produce all possible values from every hectare of forest land.  3.1 MULTIPLE USE FOREST HARVEST MODELS In order to plan for multiple use forestry, a number of models have been developed to simulate harvesting practices that could help achieve maximum benefits from a multiple use forest. This section discusses the various models that have been  40 developed to address this problem. Originally the problem was addressed only at the single stand level but was then extended to multiple stands.  3.1.1 Single stand models The earlier models developed to analyze multiple use problems were based on single stands.  The first model to account for resource values other than timber generated  by stands was developed by Hartman (1976). This is a modification of a single use timber production model originally developed by Faustman (1849), which describes the optimal economic management of an even aged stand of timber under conditions of unchanging productivity and prices over time.  Forestry literature abounds with extensions and  reformulations of the Faustman model (Newman 1988). The Hartman model assumes the flow of net benefits from timber to be a function of stand age, and identifies a rotation age that maximizes the combined present value from timber harvests and from other services provided by the standing stock. In other words, the optimum rotation age is the age when the increase in value from marginal delay of holding the stock is equal to the opportunity cost of that delay. Hartman’ s model was followed in the development of other models that examined various other aspects of multiple use (Calish et al. 1978). There is an inherent weakness in the single stand models. They assume that stands are independent of each other. But, in reality, non-timber resource values of a stand are very much dependent on its setting within the larger forest. Any activity that changes the growing conditions of an adjoining stand will positively or negatively affect one or more resource values over time. To maximize net present value from all goods and services  41 produced over time, optimal management of forests, therefore, has to consider related stands simultaneously.  3.1.2 Forest level harvest models Forest level models take into account the presence of many stands of different species and ages.  The flow of goods and services from a forest is maximized by  establishing a timber harvest pattern through time. The harvests control the flow of timber and non-timber benefits by controlling the conditions of growing stock. The amount of timber that is estimated to flow for a given period over the planning horizon is known as the timber supply. In many jurisdictions, it is through timber supply planning that the flow of non-timber resources is also generated and controlled.  Timber supply planning helps  establish an Annual Allowable Cut (AAC) for an area, which is the maximum volume of timber that can be harvested while realizing desired goals from the flow of non-timber values. In British Columbia, AAC is a policy decision after having taken into account social, economic and technical factors. Various techniques such as linear programming (LP) (Thompson et al. 1973; Davis and Johnson 1986); mixed integer programming (Kirby et al. 1986; Nelson and Brodie 1990; Nelson and Finn 1991; Murray and Church 1993), Monte Carlo integer programming and other heuristic approaches (Nelson and Brodie 1990; Dahlin and Sallnas 1993), interchange (Murray and Church 1993), simulated annealing (Lockwood and Moore 1993; Nelson and Liu 1994), and genetic algorithms (Liu 1994) have been tried to find optimal solutions to harvest scheduling problems. However, no one particular approach has been widely accepted, and the appropriate mathematical programming is still evolving.  42 3.1.2.1 Strata based models Most popular forest level harvest scheduling models are homogenous strata based models which use linear programming or simulation (Nelson and Howard 1991). They are homogenous in that each analysis unit consists of the same timber type, age class and site class (Leuscbner 1990).  In British Columbia, FSSIM which is a forest strata based  simulation model is used for harvest scheduling. The United States Forest Service uses a  linear programming based model known as FORPLAN for their timber supply planning. These models currently lack the spatial resolution needed to address specific harvesting regulations, such as maximum opening size and exclusion periods for adjacent blocks. Ignoring these factors may result in overestimating the sustainable harvest (Nelson  and Howard 1991); however, they are efficient in determining long term strategic harvest levels (Nelson et al. 1991).  3.1.2.2 Area based models Planners have recently attempted to develop spatially feasible models (Tanke 1985; Nelson and Brodie 1990; Nelson et al. 1991, Clements et.al. 1990; O’Hara et al. 1989). These models are specific to a contiguous area and are likely to be heterogeneous in that several different cover types including non forest land are included (Leuschner 1990). The decision variables in these models define treatments for specific harvest units. Since this approach requires integer solutions, optimization of this problem has been attempted using techniques such as Mixed Integer Programming, Monte Carlo Integer  43 Programming, and other heuristic approaches such as interchange, simulated annealing and genetic algorithms. There have also been attempts to link the strata based plans and area based plans (Johnson and Crim 1981; Johnson and Stewart 1987; and Nelson et al. 1991). For large forest level problems, there is currently no analytical tool available that can guarantee optimal solutions by simultaneously considering the spatial and temporal feasibility when estimating the Allowable Annual Cut (AAC).  3.1.3 ATLAS (A Tactical Land Analysis System) Model ATLAS is a simulation model developed at the University of British Columbia (Nelson et al. 1993). This spatially explicit model does not provide optimal solutions, rather it generates feasible solutions.  It simulates forest level harvests subject to spatial  constraints such as adjacency, green-up, and forest cover. Harvest units, zones and access units constitute the spatial hierarchy of the model with access units at top. Harvest priorities can be assigned to all three levels. The order of the harvest is: i) go to the top ranked access unit, ii) go to the top ranked zone within this access unit, and iii) go to the top ranked harvest unit within this zone. When all eligible blocks (harvest units) within a zone are cut, the harvest is then scheduled in the next ranked zone within the access unit. When all eligible zones within this access unit are harvested, harvest commence in the next ranked access unit. When the periodic harvest is complete (periodic harvest is satisfied or no more eligible blocks exist), harvesting stops. Blocks are then aged by one planning period and harvesting begins in the next period.  If the harvest target is not satisfied, the user must  adjust the periodic harvest targets in order to achieve the desired volume flows. As such,  44 this model can be used to examine the short term and long term effects of spatial harvesting restrictions and silvicultural treatments. It is ideally suited to examine the alternative land use planning systems proposed in this report.  3.2 LANDSCAPE PATTERN MODELING  The word “landscape” commonly refers to the landforms of a region in the aggregate (Websters New Collegiate Dictionary 1980). Forman and Godron (1986) define landscape as a heterogenous area composed of a cluster of interacting ecosystems that is repeated in similar form throughout. In this research, it refers to a spatially heterogeneous forested land surface and its associated habitats, ranging in size from a single hectare to thousands of hectares. The fundamental structural element of a landscape is a patch, which is defined as a region that is similar with respect to some attributes. Patches consist of core and edge. Core is the portion of a patch that is unaffected by neighboring patches, and usually is considered to be the interior of the patch. Edge habitat is the band on the periphery of a patch that differs abiotically from the core and may also differ biotically from the core. The spatial patterns observed in a landscape are the result of complex interactions between physical, biological and human (social) forces. Landscape patterns may respond differently to natural and anthropogenic disturbances. Since multiple use management includes maintaining visual quality and a mosaic of habitats, it is also necessary to consider landscape response to multiple use management. The forested landscapes in the vicinity of Revelstoke, British Columbia have been very much influenced by past forestry practices, and are used as a case study in this project. There have been several studies on  45  natural landscape patterns aimed at understanding the underlying ecological processes (Turner 1989; Cale et a!. 1989; Paine and Levin 1981). But studies of landscape responses to human manipulated disturbance are rare (Wallin et al. 1994). In this study, an attempt is made to simulate the landscape response to disturbance created by forest harvesting and regeneration. The extent and rate of change of the landscape pattern will reflect the availability of a mosaic of habitats over the planning horizon. Some forestry land use practices will adversely affect particular habitats while others may not. In this research I use the spatially explicit forest simulation model SIMFOR (Daust 1994) to quantify the landscape pattern response to forest harvesting and regeneration. The model uses ATLAS output of harvest schedules in a management regime and generates statistics on the resulting landscape pattern over the planning horizon. These landscape statistics are derived from the composition and pattern of ecosystem classes and seral stages. Seral stages are different ages in the succession and development of forest stands. In this study, land occupied by any seral stage that is bounded and internally homogeneous is considered to be a patch. These statistics are used to derive useful indices which help to characterize some of the long term cyclical changes in landscape composition, diversity and its biological consequences. SIMFOR helps in predicting landscape changes and in developing our understanding of the landscape dynamics associated with the likely progression of resource development in managed forested ecosystems.  46  4 EMPIRICAL STUDY  4.1 INTRODUCTION The empirical study for this research was carried out in Revelsioke Forest District. Reveistoke has severe resource use conflicts, and the management strategies developed by the District to deal with this problem are fairly advanced when compared to other districts in the province. This district has tried to design special harvesting patterns and rules to maintain a continuous flow of visual quality and wildlife habitat values.  4.2 DESCRIPTION OF THE STUDY AREA  4.2.1 General The study site is located in a sub-unit of the Revelstoke Timber supply Area (TSA), and falls under the jurisdiction of Revelstoke Forest District. Appendix Figure 1 shows the general location of the study area in British Columbia. The topography of the TSA is rugged and mountainous.  It consists of three major biogeoclimatic zones: Alpine  Tundra(AT), Interior Cedar-Hemlock (ICH) and Engelmann Spruce-Subalpine Fir (ES SF). The dominant tree species groups are mixtures of Engelmann spruce (Picea engelmanni Parry ex Engelm.) and subalpine fir (Abies lasiocarpa (Hook.) Nutt.), western hemlock (Tsuga heterophylla (Raf.) Sarg.), western redcedar (Thuja plicata Donn ex D. Don in Lamb) and Douglas-fir (Pseudotsuga menziesii (Mirb.) Franco). With such diverse biogeoclimatic zones, the region supports diverse fauna and flora.  47 4.2.1.1 Access Units The TSA has been divided into 12 Access Units or Compartments based on watershed boundaries and accessibility. This research is focussed on one of these access units, Reveistoke 1, also known as Akolkolex drainage.  4.2.1.2 Stand groups The timber supply review of the TSA (Ministry of Forests 1993a) has grouped the original forest cover types (old-growth) into nine species groups. Each of them exhibit a specific growth and yield pattern over time. It is assumed that the same species groups will be regenerated after harvest to form an additional (second-growth) nine groups, thus making a total of 18 species groups. These are called stand groups in this study. The only difference between the old-growth and the regenerated stands will be the levels of stocking and the size distribution of tree at the time of harvest. Predominant species groups in Revelstoke 1 are Douglas-fir, cedar, hemlock and spruce. The composition of stand groups and their distribution in Revelstoke 1 are illustrated in Appendix Figure 2.  4.2.1.3 Site quality Site quality is a measure of timber productivity and is an indicator of the maximum wood volume that the land can produce over a given period of time (Davis and Johnson 1987). It is commonly measured by site index which is the average total height of dominant trees at specified ages. For purposes of comparison, site indices in British Columbia are based on heights when trees are 50 years old at breast height.  For  Revelstoke, the timber supply review (Ministry of Forests 1993 a) uses two indices: i)  48 those for good-medium sites and ii) those for poor sites. The averagç. site indices for the three species considered in this research are given in Table 2.  Table 2. Site indices of selected species  @ 50 years measured at breast height (1.3m  above ground) Species  Good and medium  Poor  Douglas-fir_Interior  19  12  Spruce  18  10  Western redcedar  21  13  4.2.2 Resource use conflicts Historically, the local economy of Revelstoke was heavily dependent on resource extraction.  This, however, has changed in the recent past to include more service based  industries.  Social values have also changed dramatically.  Now there are increased  pressures to maintain wildlife and visual quality at the expense of timber production. This shift in social values in an originally timber dependent economy, with a limited land base, has led to repeated conflicts over land use issues. For management purposes, the Ministry of Forests (MOF) has classified the land base into three types: i) economically operable, ii) economically inoperable and iii) reserves. For purposes of the ATLAS analysis, these are grouped into two broad categories i) wildlands (WLD) and ii) Reserves (RSV). Wildlands refer to the productive operable forest land, while the Reserves (RSV) refer to forest lands which are economicafly  49 inoperable and where timber harvests are prohibited for environmental reasons. In the wildiands, integrated management for multiple use is practiced. Reveistoke 1 consists of 66.3 % (11649 ha) wildiands and 33.7 % (5926 ha) reserves.  This research will focus on three forest land use conflicts, namely; timber,  wildlife and visual quality.  4.2.3 Current management practices The MOF is attempting to balance the production of resource values from timber, wildlife and visual quality from the Reveistoke TSA.  This is done by practicing “total  resource planning” and management by “resource emphasis areas” (REAs). Total resource planning takes into account all resource values that have to be managed and it involves identification of not only cut blocks for harvest but also the road networks, harvest systems to be used in each cut block and other related activities. The REA designation indicates the resource value and the management objectives for a specific area within the TSA (Price and Blake 1993). Management objectives may emphasize the production of some products over others, within a multiple use framework, based on resource values and the inherent capabilities of the land.  Management practices are governed by a set of 14 Resource  Emphasis Rules (Table 3), based on the silviculture system, minimum harvest age, cut block adjacency, green-up and forest cover objectives. The rules relating to the protection of wildlife habitat are based on the requirements for ungulates. It is believed that these conditions would meet the habitat attribute requirements (Daust et al. 1993) of many other categories of wildlife.  50 Table 3. Resource Emphasis Rules for Revelstoke TSA. (Refer to Table 1 (page 8) for additional descrztion on codes). Rule No.  Resource Emphasis  Description  1  Timber: Unconstrained U  no constraints  2  Basic Timber T  adjacency and 20 year greenup, maximum distur bance rate of 40% of area, disturbance age of 20 yrs, retain 30% of area in height class 2 (40 years or older)  Note: All subsequent Resource Emphasis Rules are incremental to the Basic Timber rule Wildlife(h) WC1  retain 60% of area in height class 3 (stands 80 years or older)  Wildlife(l) WC2  retain 40% of area in height class 3  5  Wildlife(m) WC3  retain 52% of area in height class 3  6  VQ(modification) VM  maximum disturbance rate of 25%; Green-up age 40 years  7  VQ(modification) & Wildlife(h) VM-WCI  maximum disturbance rate of 25%; green-up of 40 years and retain 60% of area in height class 3  8  VQ(modification) & Wildlife(l) VM-WC2  maximum disturbance rate of 25%; green-up of 40 years and retain 40% in height class 3  9  VQ(modification) & Wildlife(m) VM-WC3  maximum disturbance rate of 25%, green-up of 40 years and retain 52% of area in height class 3  10  VQ(Partial Retention) VPR  maximum disturbance rate of 10%, green-up of 40 years  11  VQ(PR) & Wildlife(h) VPR- WC1  maximum disturbance rate of 10%, green-up of 40 years and retain 60% of area in height class 3  12  VQ(PR) & Wildlife(l) VPR- WC2  maximum disturbance rate of 10%, green-up of 40 years, and retain 40% of area in height class 3  13  VQ(PR) & Wildlife(m) VPR- WC3  maximum disturbance rate of 10%, green-up of 40 years, and retain 52 % of area in height class 3  14  Total Retention No cut TR  No logging  3  4  -  51 4.2.3.1 Resource emphasis areas (REAs) The Reveistoke TSA has four broad categories of Resource Emphasis Areas: i) timber, ii) wildlife, iii) visual quality obj ectives(VQO), and iv) VQO and wildlife. Only two REAs, wildlife and VQO-wildlife, are represented in Reveistoke 1. The distribution of REAs in the whole TSA and in Reveistoke 1 is shown in Table 4.  Table 4. Distribution of resource emphasis areas in Reveistoke 1 and the whole TSA. (Refer to Table 1 (page 8) and 3 (page 5 0) for description ofcodes). Resource Emphasis Area  TSA(ha)  Timber (7)  31,017  Wildlife  112,607  (WCJ, WC2, WC3)  Visual quality objectives VQO & Wildlife  (V VPR)  (vM-WCJ, VM-WC2,  Percent 18  Revl(ha)  Percent  0  0  63  7,924  45  4,507  3  0  0  28,814  16  9,651  55  176,945  100  17,575  100  VM- WC3, VPR- WC1, VPR- WC2, VPR- WC3)  TOTAL  Only four of the fourteen rules, (4,5,11 & 12) apply to Reveistoke 1. The rules and the areas to which they are applicable in Revelstoke 1 are listed in Table 5 which gives a breakdown of the REAs of Reveistoke 1 shown in Table 4.  52 Table 5. Summary of resource emphasis rules and their area of application in Reveistoke 1. (Refer to Tables] (page 8) and 3 (page 50) for descrztion ofcodes). Rule No.  Descrzption  Code  Area (ha)  %  4  Wildlife (low)  WC2  1324  7.5  5  Wildlife (medium)  WC3  6601  37.6  11  Visual quality partial retention and  VPRWC1  4550  25.9  VPR_WC3  5100  29.0  Total  17575  100  wildlife (high) 13  Visual quality partial retention and wildlife (medium)  53  5 RESEARCH METHODOLOGY  5.1 INTRODUCTION This chapter describes the research methodology used for the empirical study. The research was conducted in two main phases, stand level analysis and forest level analysis. Methodology for the stand level analysis consists of: i) identifying silvicultural treatments for possible inclusion in stand management; ii) defining silvicultural regimes with various combinations of silvicultural treatments; iii) simulation modeling for growth and yield of various silvicultural regimes; iv) economic analysis of selected silvicultural regimes to determine their economic feasibility at the stand level and their underlying assumptions; and v) defining basic, medium and high intensity silvicultural regimes for inclusion in the forest level analysis. The methodology for forest level analysis consists of: i) designing two alternative land use systems; ii) simulation modeling of forest level timber harvesting in the two systems; iii) simulation modeling of landscape pattern responses to forest level harvesting in the two systems; and iv) economic analysis of forest level harvesting and its underlying assumptions.  Finally, some economic and  environmental parameters that could possibly be used as indicators of net benefit to society are identified.  5.2 TOTAL RESOURCE PLANNING Prior to discussing methodology it is necessary to understand the planning process for multiple use management in the Revelstoke Forest District.  This is done through a  process called “total resource planning” which identifies all resource values, both local (e.g.  54 timber, critical wildlife habitat, hydrologically sensitive areas, visual quality etc.) and global (e.g. biodiversity), and develops long term management strategies to maintain and enhance them. This aids in the achievement of multiple use objectives and also avoid the dangers of piecemeal development of certain areas for specific uses. The Reveistoke Forest District has already developed such a plan for the Reveistoke TSA. The following data pertaining to the plan for Reveistokel was supplied by the Ministry of Forests: digitized data for the total resource plan for Reveistoke 1, showing Resource Emphasis Areas, harvest cut blocks (or polygons), and road networks; •  resource emphasis areas and the resource emphasis rules applicable to them;  •  harvest system used for each cut block; and  •  costs for harvesting, tree-to-truck, hauling, road construction and maintenance.  5.2.1 Temporal Modification of Forests Past management practices relied on natural regeneration augmented by artificial regeneration which generally resulted in the regeneration of the original species groups. Under current management practices, however, all harvested areas are artificially regenerated with selected species. In this research, it is assumed that: •  all existing regenerated species (second-growth) are similar to the original species except for the reduction in areas (11% of the area) resulting from roads and skid trails (Ministry of Forests, 1993a), and  •  all future harvest areas will beregenerated with Douglas-fir, spruce or cedar (which are the predominant species planted).  55 The conversion of old-growth and regenerated species groups to modified stands generally follows the pattern advocated in Table A14 of the Reveistoke TSA Timber Supply Review (Ministry of Forests 1993a) where a specific mixture of species replaces a particular species group. In this research, to facilitate growth and yield modeling, it is assumed that instead of regenerating a mixture of species, the dominant species in the group is regenerated as a pure crop. The conversion pattern is given in Appendix Table 2. Using these assumptions, this research distinguishes the following three classes of stand groups: •  old-growth stands that have never been harvested or affected by natural calamities,  •  second-growth stands with the same species composition as the old-growth stands, and  •  modified stands with selected tree species.  5.3 METHODOLOGY FOR STAND LEVEL ANALYSIS  5.3.1 Silvicultural treatments Silvicultural treatments are techniques used to increase the quantity and quality of timber produced from fixed units of land. Silvicultural treatments generally involve a high ratio of capital and labor input to a fixed unit of land. A description of the silvicultural treatments considered in this research are given below.  5.3.1.1 Artificial regeneration (P1) 1600 stems per hectare (sph) are planted within one year of regeneration delay is assumed.  harvest.  No  56 5.3.1.2 Pre-commercial thinning (PCT) The objective of pre-commercial thinning is to attain the greatest possible residual tree diameter with the least amount of input (Smith 1986). Trees are pre-commercially thinned when they are approximately 4m in height. Trees are removed so that the resulting stand retains well spaced, large diameter at breast height (DBH) trees.  Two types of PCT  are distinguished, based on the residual crop density: i) 1200 stems per hectare (sph) and ii) 800 sph.  5.3.1.3 Commercial thinning (CT) Worthington and Staebler (1961) define commercial thinning as thinning programs that produce merchantable products that have a value equal to or greater than the cost of extraction.  Smith (1986) explains thinning as a way of allocating production to some  optimum number of trees of highest potential to increase value: while removing the other trees systematically in such a sequence as to obtain the maximum economic advantage. It is not clear from the literature as to what should be the ideal residual stock that will maximize returns as this varies considerably with species and site. The essence of good thinning practice is not only to realize a present profit, but also leave the stand in a condition that will enhance future returns.  Stone (1993) in his study of commercial thirinings in coastal  Douglas-fir (site indices 36, 30 and 24) used 200, 300, 400, 500, 600 and 800 sph as post thinning densities. In this research, commercial thinning was only considered for stands growing on good and medium sites. Thinning was scheduled when half of the rotation age volume is  57 attained. At the time of thinning, one of the following two options was applied: i) retain  two thirds (2/3) volume in the residual crop, and ii) retain one half (1/2) of the volume in the residual crop.  These generally resulted in post-thinning densities of 800 sph and 600  sph respectively.  5.3.1.4 Pruning (P) The main objective of pruning is to improve the quality of wood by increasing the proportion of clearwood in the timber. Pruning was applied to stands growing on good and medium sites. Two pruning lifts were modeled.  The first lift was up to 3m height and it  was done at an average stand height of 6m. The second lift was up to 5.5m height and was done at an average stand height of 1 0.5m. All trees that meet the requirement of retaining 50% of the live crown were pruned. This amounts to roughly fifty percent of the standing crop.  5.3.1.5 Fertilization (F) Fertilizer application was modeled only for Douglas-fir stands because the TASS model (discussed in Section 5.3.2.1) that was used for stand growth simulation can simulate fertilizer application only for this species (Polsson 1994). Fertilizer was applied at a rate of 225 kg per hectare. Two applications of fertilizer (first at 4m height and second at 6m height) were modeled.  5.3.2 Silvicultural regimes Silvicultural regimes refer to a series of silvicultural treatments applied to a stand over its life. In this research, an attempt is made to push timber production towards its  58 intensive margin by employing three different silvicultural regimes namely, basic, medium and high. These intensity levels based on their respective levels of investment. The broad category of treatments that were included for each of the above silvicultural regimes were decided a priori based on Ministry of Forests experience and from personal observation. Specific treatments to be included in each silvicultural regime was done after an economic analysis discussed in Section 5.3.4. The broad category of silvicultural treatments included in the three silvicultural regimes are given below: •  Basic: Natural regeneration of stands  •  Medium:  Artificial regeneration and pre-commercial thinning (PCT). Only stands  growing on good and medium sites were treated with PCT. •  High: Artificial regeneration, fertilization, pruning, and commercial thinning. Fertilization, pruning and commercial thinnings were only applied to stands growing on good and medium sites.  Due to difficulties encountered in predicting the growth  response of various species to fertilization, it was used only for Douglas-fir. A total of 135 silvicultural regimes formulated by a combination of silvicultural treatments (planting, pre-commercial thinning, pruning, fertilization, and commercial thinning) were modeled on TASS by the Ministry of Forests (MOF). Table 6 shows the number of silvicultural regimes developed for each species and the type of silvicultural treatments examined in each regime. Details are given in Appendix Table 3.  59 Table 6 Silvicultural treatments examined for development of silvicultural regimes. (Refer to Table ifor descrzption ofcodes). Spp SI  Regimes  Silvicultural treatments  (total) PC8  PC5  P1  P2  (sph) (sph)  (sph)  (sph)  @6m  @lOm @4m  @50 P1  PCI2  Fl  F2  CT  CT  @6m  (112)  (113)  Df  19  1600 yes  yes  yes  yes  yes  yes  yes  yes  yes  54  Df  12  1600 yes  yes  yes  yes  yes  yes  yes  no  no  43  Se  18  1600 yes  yes  no  yes  yes  no  no  yes  yes  13  Se  10  1600 yes  yes  no  yes  yes  no  no  no  no  6  Cw  21  1600 yes  yes  no  yes  yes  no  no  yes  yes  13  Cw  13  1600 yes  yes  no  yes  yes  no  no  no  no  6  Total number of regimes  135  5.3.2.1 Simulation modeling for growth and yield Timber growth and yield refers to the prediction of growth and development of individual stands in response to management inputs over time. Timber volumes for the oldgrowth and naturally regenerated stands are based on the variable density yield prediction (VDYP) model developed by the B.C. Forest Service Inventory Branch (Ministry of Forests 1993b). VDYP is an empirical yield prediction system for natural stands. It provides estimates of merchantable volumes for existing and regenerated stands according to their composition and age. Timber volumes for the modified stands  are generated by the Tree and Stand  Simulator (TASS) model. TASS is a biologically oriented model designed to assess the  60 effects of cultural practices and environmental factors on the growth and yield of forest tree species (Mitchell 1975). In using these models, it is assumed that the effects of irregular stocking, pests and other factors that contribute to mortality have been compounded into the empirical calibration of the model. The growth and yield of Douglas-fir, western redcedar and Engelmann spruce were modeled with respect to two site indices using TASS.  Descriptions of the types of  treatments used are discussed in Section 5.3.1. 5.3.2.2 Simulation modeling for bucking and sawing Bucking and sawing of logs to specific sizes was necessary to estimate the value of wood of different sizes and quality to carry out economic analysis. Bucking was simulated by means of a program custom made for this study by the MOF using the data generated by the TASS model.  This program groups pruned and unpruned logs  separately. The harvested timber up to 10 cm top diameter was bucked into 5.3 m logs and sorted into five top diameter classes (less than 10 cm, 10-19 cm, 20-29 cm, 30-39 cm, 40-49 cm). The top end of the timber below 10 cm in diameter was taken as pulpwood. To estimate the premium on pruned logs, sawing was simulated on pruned logs using the SAWTAB program in the SAWSIM model developed by Halco Software Systems Ltd. This program is used to maximize the value of lumber from a particular log given the values of different types of boards. In this study, this’ program was used only to produce 2 by 2 squares from pruned logs and to sort them into clear and knotty lumber,  61  5.3.3 Selection of rotation age An important variable in any analysis of silvicultural operation is the rotation age. Selection of a rotation age for timber depends on management objectives as to what is to be maximized, whether it is a specific technical product, biological yield or return on investment.  Accordingly, these are termed technical, physical and economic rotations,  respectively. The MOF mostly uses physical rotations in their estimation of AAC. In this research, I also use the physical rotation age (which is the age at which the mean annual increment culminates). This is applicable for all species regenerated in the modified forests. When intensive management is practiced on a particular species, the physical rotation age is either shortened or extended. Ideally this research should use different physical rotation ages for different silvicultural regimes. Considering the limited time and other resources available, it was necessary to assume that the rotation age for a particular species (with a specific site index) will remain constant irrespective of the silvicultural regime. Determining optimal rotation ages at the stand level is a complex problem in its own right, and this is not an objective of this study. Clearly, the stand level optimization of value ultimately needs to be incorporated into the decision making frame work developed here. Since the old-growth consists of a mixture of species of varying age classes, it is not possible to define rotation ages specific to each species. They are generally over 120 years and are ready for harvesting at any time.  For regenerated stands, depending on the  productivity of the site, and species, rotation ages varying from 100 to 140 are used. These rotation ages are the same as those used by the MOF in its recent Timber Supply Review (Ministry of Forests 1993a) but are probably conservative for managed stands.  62  5.3.4 Stand Level economic analysis In British Columbia, stand level economic analysis is not normally used for making decisions. However, for purposes of this research, those treatments were chosen which will yield a rate of return to capital investment of 3% or more for a single rotation. Based on the information obtained from growth and yield analysis of the silvicultural regimes identified in Table 6 in Section 5.3.2, fifty four (54) silvicultural regimes were selected for stand level economic analysis. This was done to determine their economic feasibility and ranking according to the following criteria.  i)  Affordable cost per treatment This is the value of discounted total revenue at 3%. As long as the cost of a  particular treatment (discounted at an annual rate of 3%) is less than this value, the treatment will generate positive returns.  iz) Actual costs of treatments These values were obtained (when available) from the MOF. Since many of these treatments have never been implemented, it is difficult to obtain experienced costs. Therefore, in order to generate these figures, information has been pooled from various sources within the Ministry of Forests, from licensees and from the author’s own experience.  iiz2 Discounted Net Revenue This is the difference between discounted total revenue and discounted costs of a treatment. When this figure is negative, the treatment is not feasible. Higher values imply higher returns.  63 The treatments that were found to be economically feasible and that gave the highest net discounted revenue were considered for inclusion in the silvicultural regimes of intensive timber management. 5.3.4.1 Assumptions used in the stand level economic analysis The following major assumptions were used for the stand level analyses. i)  Silvicultural treatments except pruning are assumed to affect only the diameter and height of the trees and not the quality of the wood. Pruning increases the clearwood in the timber.  Ii;)  The total volume in modified forest is considered merchantable because the lower diameter classes (which cannot be utilized as lumber) are utilized as chips.  iii) All prices are current.  Logs of different diameter classes belonging to different  species were priced as per prevailing prices in the Interior.  Prices used are the  average of the prices given by Downie Street Mills at Reveistoke and by Federated Cooperatives Ltd. of Salmon Arm. The class which is less than 10 cm top diameter is considered suitable for pulpwood (chips). An examination of the selling price index for British Columbia softwood lumber over the last two decades indicates that the last price cycle was during the period 1986 to 1994. The current prices appear to fall on the peak of the cycle. Price sensitivity analysis based on long term trend was done for forest level analysis. Prices used in the analyses are given in Appendix Table 4.  64 iv) When pruning is considered as a treatment, the logs are sorted into two broad sorts, primed and unpruned. Pruned is sorted into 4 diameter classes excluding the lowest diameter class which constitutes the chips, while unpruned is sorted into 5 diameter classes including the lowest diameter class. v)  It is extremely difficult to estimate the price of pruned logs as pruning has never been done on a management scale in the Interior. Mitchell et al. (1989) state that clear lumber could bring in a premium of 3 00%. For the purpose of this analysis, I estimated the value of pruned logs as follows. First, estimate the percent of clear lumber in pruned logs of different diameter classes by comparing logs from a treated stand with an untreated stand. Then increase the price of that percent of clearwood per log class by 300 percent, while using the normal price for the knotty portion of the log. Price increases vary slightly with species and time of harvest. In the interest of time and simplicity, I have assumed that this percentage is constant across species and time of harvests (i.e., I have used the Douglas-fir premiums for the other two species).  vi) Real price increases of 0.1% to 1% per annum were assumed depending on species and diameter (Appendix Table 5). These price trends were obtained from Laing & McCulloch (1993). vii) Clear felling cost is assumed to be $23/rn , while commercial thinning cost is 3 3 (Nelson 1994). assumed to be 25% higher at $ 28.75/m viii)  Hauling cost is assumed to be $7.65/rn . 3  ix)  A real interest rate of 3 % was assumed.  65 x)  Administrative costs are assumed to be constant across all treatments.  5.4. METHODOLOGY FOR FOREST LEVEL ANALYSIS  5.4.1 Alternative land use systems Having completed the stand level analysis which consists of designing suitable silvicultural regimes for inclusion in the forest level model, the next stage is to define alternative land use systems to be analyzed. Two types of multiple use land use systems, referred to as the Integrated Use (IU) system and the Single Use (SU) system, were devised. The IU system treats the whole operable area as an integral production unit where timber is produced as one of the multiple uses, with all spatial and temporal constraints in place. This system requires that each hectare be managed for multiple use according to the appropriate resource emphasis rules. In the SU system, a portion of the operable area is allocated as a single use area, specifically for timber production. The balance of the area is managed for production of multiple uses (wildlife and visual quality) other than timber. The timber production zone of the single use system is the minimum area required to produce an even-flow volume, equivalent to the maximum even-flow achievable with current management practices from Revelstoke 1, under unconstrained conditions. This is determined by several iterations involving the systematic reduction of the number of zones.  5.4.2 Simulation modeling for forest level harvesting The spatial timber supply model ATLAS version 1.3 (Nelson et al. 1993) was used for harvest simulation of the forest.  Harvest cut blocks are assigned as harvest units of  ATLAS and different resource emphasis rules are then assigned to zones to simulate  66 resource emphasis areas. Harvest priority was based on the minimum distance from the start of the road network. This “most accessible block first” priority minimizes the amount of road construction which is an important consideration in spatially constrained harvest scheduling (Nelson et al. 1994).  5.4.2.1 Planning horizon Harvest simulations are analyzed for two planning horizons (PH); i) long term (120 years approximately one rotation), and ii) very long term (240 years approximately two -  -  rotations). The very long term is explored so that the forest reaches a steady state, where all, or most, of the yields are from regenerated stands. Planning periods are defined as 10 years and 20 years for long and very long term, respectively. Harvest flows are constrained by an even-flow policy. Evenflow policy was adopted because i) the forest planning in British Columbia is volume based and ii) sustained yield is the premise on which cut regulations are based (Williams 1993). This is a departure from the current MOF practice which determines the annual allowable cut (AAC, the volume to be harvested each year) administratively by taking into account the Long Run Sustained Yield (LRSY), and other social, economic and environmental factors.  5.4.2.2 Parameters determined by ATLAS simulation modeling The following attributes for each planning period were determined by means of ATLAS simulations: •  even flow volume;  •  area harvested;  67 •  harvest system costs;  •  length of roads constructed and maintained and the respective costs;  •  delivered wood costs; and  •  ratio of length of edge to total area.  5.4.2.3 Harvest scenarios Harvest scenarios that were designed can be grouped into three broad categories: i) current management practices, ii) alternative land use systems, and iii) enhancement of nontimber values within the timber zone. These are described in the following sections.  5.4.2.3.1 Scenario modeling current managementpractices Current management practice with all resource emphasis rules in place was modeled.  Having this as the base case scenario, other scenarios were developed by  systematically relaxing each of the resource emphasis rules. This helps to estimate the impact of each of these rules in terms of timber volume.  5.4.2.3.2 Scenario modeling alternative land use systems Economically feasible silvicultural regimes at basic, medium and high intensities that yielded the highest net discounted revenue (as per stand level economic analysis discussed in Section 5.3.4) were used to define three intensities (basic, medium and high) of timber management at forest level. These three intensities of timber management were used for forest level harvest simulation of both the SU and the IU systems. A summary of the harvest scenarios simulated are given in Table 7.  68 Table 7  Harvest scenarios in the SU and the IU systems with intensive timber  management. (Refer to Table ifor description ofcodes). Single Use  Integrated Use  timber management  (SU)  (IU)  Basic  SU_1  lU_i  Medium  SU_2  IU_2  High  SU_3  IU_3  Intensity of  5.4.2.3.3 Scenario modeling enhancement of non-timber values within the timber zone This scenario was modeled to determine the impact of various harvesting constraints on intensive timber management. In this case, the single use area designated for timber production is treated as an independent unit. Fourteen other scenarios were developed from the base case scenario of unconstrained timber production at basic intensity by including intensive timber management and by systematically introducing of wildlife and visual quality constraints. The harvest scenarios developed are given in Table 8. Table 8 Harvest scenarios for the timber zone with enhanced non-timber values.  (Refer to Table 3 & 7 for description of codes. First part of the code refers to land use system (Table 7) and the secondpart refers to resource emphasis rule (Table 3)) Single Use  Wildlife_basic  Wildlife_medium  VQ_basic  VQ_medium  (SU_U)  (‘I’)  (WC1)  (YM)  (VPR)  Basic  SUI  SUiT  SU1WC1  SU1VM  SU1VPR  Medium  SU_2  SU2_T  SU2_WC 1  SU2_VM  SU2_VPR  High  SU_3  SU3_T  SU3_WC1  SU3_VM  SU3_VPR  Intensity  69  5.4.3 Simulation modeling for landscape pattern responses SIMFOR was used to model the response of landscape patterns to intensive management practices for both the SU and the IU systems over a 240 year planning horizon. The even-flow harvest schedules generated by ATLAS simulations for each of the management option were used as input data to SIMFOR. generates a set of landscape statistics.  The SIMFOR model  From these landscape statistics, the following  landscape pattern indices were derived. z)  Distribution of seral stages over the 240 year planning horizon. The seral stages used in this study are given in Table 9.  ii)  Types of ecosystems represented within various seral stages over the 240 year planning horizon.  Three biogeoclimatic zones, the Moist Warm Interior Cedar  Hemlock Zone (ICHmw), Wet Cool Interior Cedar Engelmann Spruce  -  -  -  Hemlock Zone (ICHwk), and  Subalpine Fir Zone (ESSF) are recognized within the  management unit (Meidinger and Pojar 1991; Braumandl and Curran 1992). Detailed ecosystem mapping complete with biogeoclimatic subzones and their variants were not available for this study. iii)  Percent of area within old-growth (121  -  240 years) and very old-growth (>240  years) that constitute the edge habitats. The width of the edge band is assumed to be i) 100 meters for contrasting patches of regeneration seral stage and old-growth or very old- growth, and ii) 50 meters for contrasting patches of pole seral stage and old-growth or very old-growth seral stages. This is based on the premise that the edge influence extends one or two times the height of the trees (Bradshaw 1992).  70  iv) Percent of area within the regeneration seral stage that is affected by adjoining oldgrowth. The width of the regeneration edge affected is assumed to be 100 meters for contrasting patches of regeneration and very old-growth (> 240 years) seral stages, and 50 meters for contrasting patches of regeneration and old-growth (120  -  240  years) seral stages.  v)  Patch sizes and their distribution in very old-growth (>240 years) during a 240 year planning horizon. Patch sizes used in this study are: a) patch 1  =  0  b) patch 2  =  101  c) patch 3  =  501  d) patch 4  -  -  -  100 ha; 500 ha; 1000 ha; and  greater than 1000 ha.  Table 9 Seral stages distinguished in the land use systems and their description Seral stage  Age class (years)  Description  seral 1  0  20  regeneration  seral2  21  60  pole  seral 3  61  seral 4  121- 240  old-growth  seral 5  >240  very old-growth  -  -  -  120  mature  544 Forest level economic analysis The forest level economic analysis was carried out for all harvest scenarios developed in section 5.4.2.3. The following economic parameters were determined.  71 z) Total revenue This is revenue earned from harvested timber in each planning period.  iz) Delivered Wood Costs This includes the following phase costs: tree to truck; road construction; road maintenance and hauling.  iii)  Administrative costs These are the public costs incurred for administration, planning and  implementation.  iv) Timber rent This is the return to the productive capacity of the land under timber management. It is approximated in this study by summing the net present values for all planning periods. 5.4.4.1 Assumptions used in the forest level analysis The assumptions used are essentially the same as those for the stand level analysis. Additional and modified assumptions are listed under two categories i) general and ii) economic. 5.4.4.1.1 General assumptions z)  The forests are publicly owned and publicly managed.  ii)  There is no loss in site productivity with the second and third rotation crops for both the IU and the SU systems.  72 iii)  Watershed management practices are assumed to be the same in both the IU and the SU systems and, therefore, water quality is assumed to be the same under both systems.  iv) Natural calamities such as fire will not affect the forest management unit during the planning period. 5.4.4.1.2 Economic assumptions i)  A real interest rate of 2% is assumed (sensitivity to interest rates of 0% and 4% were carried out later).  ii)  Commercial thinning cost is assumed to be 25% higher than that of clearfelling. The . 3 cost of commercial thinning is estimated to be $28.75 /m  iii)  Analysis is carried out with and without real price increase assumptions.  iv) For purposes of simplification average price was used for all species and price distinctions were not made among species. v) The average price of logs of different diameter classes is obtained by weighting the volumes produced by each species in one rotation (Appendix Table 6). vi) The cost of silvicultural treatments (compounded or discounted appropriately) are deducted from final crop revenues and not from the thinnings, as the return to investment occurs only with the final crop. vii) At the forest level, timber may not always be harvested at the exact rotation ages prescribed by stand level analysis. As more and more resources are managed, within a limited land area, harvesting of some of the stands often gets postponed. This is  73 likely to either increase or decrease returns depending whether the delay moves the harvest age towards or away from its economic rotation age. viiz)Administrative costs consist of two categories: i) management and protection, and ii) planning, implementation and monitoring. They are estimated at $6.76 and $8.22 per hectare, respectively. These estimates were compiled by using budgeted expenditures for these programs in Revelstoke Forest District for the year 1993. For the IU system, both costs are applied to the gross area of Revelstoke 1.  For the SU system,  management and protection costs are applied to the gross area of Revelstoke 1 while the planning, implementation and monitoring costs are applied to the timber zone only (as there are no timber management activities outside of the timber zones).  5.4.4.2 Price sensitivity analysis Price sensitivity analysis was carried out with increases and decreases in current prices by 12%. This was arrived at by examining selling price indices for B.C. Interior softwood lumber for the period 1981 to 1994. On the assumption that the last price cycle was from 1986 to 1994, the average price increase was estimated to be 12% above the base year 1986=100. Price sensitivity analysis was done for the IU and SU systems at basic, medium and high timber production intensities. 5.5 INDICATORS OF NET BENEFIT TO SOCIETY  The following economic and environmental parameters determined for both the IU and the SU systems were used as measures or indicators of net benefit to society. Economic parameters reflect the contribution to the material well-being of society, while  74 the environmental parameters reflect the impact of multiple use management on ecosystem integrity. The parameters are determined for a 240 year planning horizon.  5.5.1 Economic parameters i)  Opportunity cost of resource emphasis rules, measured in terms timber supply and rent forgone over a period of 120 years.  ii)  Timber rent for a 240 year planning horizon, measured in dollars.  iii)  Periodic (20 years) cost of construction and maintenance of roads measured in dollars over a 240 year planning horizon.  5.5.2 Environmental parameters i) Road density measured in terms of average length of roads (in kms) maintained for each 20 year period and the length of road (in kms) opened during a 240 year planning horizon. ii) Periodic (20 years) distribution of seral stages of timber expressed as percent area of  total land base. iii)  Biogeoclimatic zones represented within the very old seral stage, expressed as percent area of total land base per 20 year period, over a 240 year planning horizon.  iv) Periodic (20 years) distribution of edge habitats available in the landscape expressed as percent area of the very old-growth. v) Regeneration area affected per 20 year period by edges of the old and very old seral stages, expressed as percent area of regeneration per period.  75 vi) Periodic distribution of patch sizes within the very old seral stages of timber,  expressed as percent area of the total land base.  76  6 STAND LEVEL ANALYSIS  6.1 INTRODUCTION This chapter gives the results of the stand level growth and yield analysis and the economic analysis for the six stand groups (Table 2 in page 48) that constitute the modified forests. It also discusses the relevance of stand level analysis to the forest level analysis.  6.2 ANALYSIS OF GROWTH AND YIELD  6.2.1 Growth patterns of stand groups Growth curves of all six stand groups were modeled using TASS. Growth curves of three species: Douglas-fir (SI= 19), cedar (SI=21), and spruce (SI=18) are illustrated in Figure 8 which shows that cedar consistently produces a higher volume than spruce and Douglas-fir.  Spruce produces higher volumes than Douglas-fir between 60 and 180  years. Patterns of growth exhibited by stands subjected to silvicultural treatment regimes were also modeled.  Figure 9 illustrates the pattern of growth with pre-commercial  thinning and commercial thinning in Douglas-fir stands (SI=19). It shows that stands subjected to PCT converge with the untreated fully stocked stand around 180 years, but the growth in stands subjected to CT do not converge. The time it takes to converge with the growth curve of the untreated stand is dependent on the type, intensity and timing of the silvicultural treatments. Treatments such as fertilization and genetic improvement will not only converge early but may also shift the growth curve upwards.  This  77 information is very important in interpreting even flow volumes with respect to the areas harvested and in planning area-based management of forests.  6.2.2 Effect of silvicultural treatment regimes on age of maximum mean annual increment (MAI) in volume Silvicultural treatments increase or decrease the age of maximum MAT of a stand. These effects of silvicultural treatment regimes on maximum MAIs were determined for the six stand groups.  Results showing the effect of pre-commercial thinning (PCT),  commercial thinning (CT) and fertilizer applications on maximum MAI of Douglas-fir (SI=19) are illustrated in Figure 10. While PCT and CT increase the age of maximum MAI, applications of fertilizer decrease it.  6.2.3 Determination of rotation ages Rotation ages of any stand based on the maximum MAT will vary with the type of silvicultural treatment. This study uses the maximum MAT of the untreated stand of each of the six stand groups as its rotation age, irrespective of the type of silvicultural treatment regimes they are subjected to. The rotation ages of the six stand groups of the modified forest as determined by the growth and yield analysis are given in Figure 11.  78  1600 1400  :  .  1200 1000  E 0  __ fi 9 _c21  800  __•__ sI 8  E 600 400 200 0  10  20  30  40  60  50  80  70  90  100 110 120 130 140 150  160 170 180 190 200  Stand age (years)  Figure 8 Growth curves for Douglas-fir (SI=19), cedar (S121) and spruce (S118). (Refer to table 1 (page 8)for descr4’tion ofcodes).  800 700 600 N  ..  500 ..  E 0  E  —A—  E  NOTREAT PC8 CT(1/2)&F2  100  .  +  10  20  30  40  50  .  60  70  80  90  100 110 120 130 140 150 160 170 180 190 200  Stand age (years)  Figure 9 Effect of silvicultural treatments on growth curves of Douglas-fir (SI=19). (Refer to Table 1 (page 8)for descrztion ofcodes).  79  140 120 100 80 >. Q  60 40 20  -  0Silvicultural regimes  Figure 10 Effect of silvicultural treatment regimes on age of maximum MA! of Douglas-fir (S119). (Refer to Table 1 (page 8)for description ofcodes).  80 0  60 40 20 0 Stand groups  Figure 11 Rotation ages of stand groups used in the second-growth forest. (Refer to (Table 1 (page 8)for description ofcodes).  80  6.2.4 Effect of silvicultural treatment regimes on volume and diameter at breast height (DBH) The effect of silvicultural treatment regimes on volume and DBH of the fmal harvest varies with stand groups. In this study, since the final harvest of all stand groups takes place at the culmination age of an untreated stand, depending on the type of treatment there are varying degrees of compromise between the volume harvested and the increase in DBH achieved.  These effects for the six stand groups are illustrated in  Figures 12 through 17. These figures show that, in all cases, there is loss in volume and increase in DBH. In the case of Douglas-fir, on both good and poor sites, it is possible to make up for the loss in volume while increasing the DBH through the application of fertilizers.  35 30 25 20  E 0  15  E z  ToIume __dbh  J  0  >  10 5 0 PC8  CT(1/2)  CT(1/2)&F2  Silvicultural treatments  Figure 12 Effect of silvicultural treatment regimes on Volume and DBH of final harvest in Douglas-fir on good and medium sites (SI=19). (Refer to Table 1 (page 8) for description of codes).  81  25  210 200  ‘  20  190 15 oI@120  180  X  E  _._dbh@120  io 170 5  160 150  0• NOTREAT  PC8  PC12  PC12&F2  Silvicultural treatements  Figure 13 Effect of silvicultural treatment regimes on Volume and DBH of final harvest in Douglas-fir on poor sites (S112). (Refer to Table 1 (page 8) for descrzption ofcodes).  900  45  800  40  700  35  600  30  500  25  400  20 x  300  15  200  10  100  5  V01 _._dbh  0  0 NoTreat  PC8  CT(1/2)  Silvicultural treatments  Figure 14 Effect of silvicultural treatment regimes on Volume and DBH of final harvest in redcedar on good and medium sites (S121). (Refer to Table 1 (page 8)for description ofcodes).  ______  ___________  82  30  420 410  25 400 20  390 380  VoI  15  _._dbh  370 10  360 350 I  340  !;..:  330 NOTREAT  5 0  PC8  PC12 Silvicultural treatments  Figure 15 Effect of silvicultural treatment regimes on volumes and DBH of final harvest in redcedar on poor sites (S113). (Refer to Table] (page 8) for descrztion of codes).  35  530 520  30  510 25  500 •••  490 480 •.•  .  •1  20 E  U  VoI  x  _._dbh  15  470  470  10  460  I.  450  5 0  440 NOTREAT Silvicultural treatments  Figure 16 Effect of silvicultural treatment regimes on volume and DBH of final harvest in spruce on good and medium sites (SI=18). (Refer to Table] (page 8) for description ofcodes).  83  520 510 500 20 E  490  [L 15  480  .  VoI .dbh  470 460 450 Silvicultural treatments  Figure 17 Effect of silvicultural treatment regimes on volumes and DBH of final harvest in spruce on poor sites (SI=10). (Refer to Table 1 (page 8) for description of codes).  6.3 STAND LEVEL ECONOMIC ANALYSIS  Stand level economic analysis was carried out for 54 selected silvicultural regimes from the six stand groups, based on the information obtained from the growth and yield analysis. The selected silvicultural regimes are given in Tables 10 and 11. Table 10 consists of treatments carried out in all six stand groups. Table 11 consists of additional treatments carried out in Douglas-fir, because application of fertilizer could be modeled only for Douglas-fir. Premiums on various sizes of pruned logs was estimated by first determining the percent of clear wood in various diameter classes of Douglas-fir pruned logs by using  84 TASS. The results of this analysis are illustrated in Figure 18. As expected, a higher proportion of clear wood can be seen in larger diameter classes. Then, the prices of logs in each diameter class were increased by a premium of 300% on the percent of clear wood in that diameter class. Results showing the percent increase in the value of pruned logs of different diameter classes in Douglas-fir are shown in Figure 19. The results show pruned log premiums range from 49% for the lowest diameter class to 114% for the highest diameter class. Tables 12 and 13 summarize the results of the economic analysis indicating whether a particular treatment is economically feasible or not.  Results of the economic  analysis for redcedar from good and medium sites are illustrated in Figures 20 and 21. The results of other stand groups are illustrated in Appendix Figures 3 through 14. These figures clearly show the ranking of treatments with respect to their net discounted revenues (3% discounted rate).  Table 10 Silvicultural treatment regimes selected for stand level economic analysis. (Refer to Tables ](page 8) and 6 (page 59) for descrztion ofcodes). Species Silvicultural treatment regimes CT(113)_P2  CT(112)  CT(112)_P2  yes  yes  yes  yes  yes  yes  yes  yes  yes  yes  yes  yes  yes  yes  jyes  yes  yes  yes  no  no  jno  no  yes  yes  yes  yes  no  no  no  no  yes  yes  yes  yes  no  no  PCI2  PCI2_P2  PC8  PC8_P2 CT(113)  f19  yes  yes  yes  yes  c21  yes  yes  yes  sIB  yes  yes  f12  yes  c13 slO  no  85  Table 11 Additional silvicultural treatment regimes selected for Douglas-fir (S119 &12). (Refer to Tables 1 (page 8) and 6 (page 59) for description of codes). Spp. Silvicultural treatment regimes PCI2 PCI2 PCI2  PC8 •PCBPC8  iCT(1!3) ICT(1/3) CT(113) ICT(112)CT(112)ICT(112)  _F1  _F2  _F2P2 _F1 _F2 ;_F2P2 _F1  _F2  _F2P2  _F1  _F2  _F2P2  119  yes  yes  yes  yes yes yes  yes  yes  yes  yes  yes  yes  112  yes  yes  yes  yes yes yes  no  no  no  no  no  no  60  a  0  20_29 Diameter classes (in cms)  Figure 18 Percent of clear lumber in pruned logs of different diameter classes in Douglas-fir (S119) that is commercially thinned @70 years and harvested @130 years.  ______  86  120 100 S  80  S  S 0 S S U C  60 40 20  0 30_39  20_29  10_19  40_49  Diameter classes (in cms)  Figure 19 Percent increase in value of pruned logs of different diameter classes in Douglas-fir (SI=19) that is commercially thinned at 70 years and harvested @130 years.  8000 7000 6000 S 44  5000  Aff$  0,  0 U  S  E  Act$  4000 3000  (0 S  I-  2000 1000 0 PCI2  PC8  CT(1/3)  CT(1/2)  PCI2_P2  PC8_P2  CT(1/3)_P2 CT(1/2)_P2  Treatment Type  Figure 20 Redcedar (SI=21): Actual and affordable costs (with real price increases) for selected silvicultural treatment regimes. (Refer to Table 1 (page 8) for descrztion of codes).  87  4500 4000 3500 3000  ,  -  -  2500 2000  I  1500 10000  500 0-  Treatment Type  Figure 21 Redcedar (SI=21): Feasibility as indicated by discounted net revenues (with real price increases), of selected silvicultural treatment regimes. (Refer to Table 1 (page 8)for description ofcodes).  Table 12 Silvicultural treatment regimes showing economic feasibility. “Yes” refers to feasible regimes. (Refer to Tables 1 (page 8) and 6 (page 59) for description of other codes). Spp. Silvicultural treatment regimes PCI2 PCI2_P2 PC8  PC8_P2 CT(1!3)  CT(113)_P2 CT(112)  CT(1!2)_P2  f19  no  no  no  no  yes  yes  yes  yes  c21  yes  yes  yes  yes  yes  yes  yes  yes  sIB  no  yes  yes  yes  yes  yes  yes  yes  f12  no  no  no  no  not done not done  not done  not done  c13  no  no  no  no  not done not done  not done  not done  sIC  no  no  no  no  not done not done  not done  not done  88 Table 13 Additional silvicultural regimes for Douglas-fir showing economic feasibility. “Yes” refers to feasible regimes. (Refer to Tables 1 (page 8) and 6 (page 59) for descrztion ofother codes). Spp.Silvicultural treatment regimes  ;PCI2 PCI2 PCI2 PC8 PC8 :PC8 Fl  F2  F2P2  Fl  F2  CT(113) CT(113) CT(1I3)CT(1I2) CT(112) CT(112)  F2P2 Fl  F2  F2P2  Fl  F2  F2P2  f19  no  no  no  no  no  no  yes  yes  yes  yes  yes  yes  f12  no  no  no  no  no  no  not  not  not  not  not  not  done  done  done  done  done  done  6.4 IMPLICATIONS FOR FOREST LEVEL ANALYSIS  Stand level analysis has shown that the rotation age of any species will vary with site productivity and with type, age and intensity of silvicultural treatments. Selection of a single rotation age for any species for any silvicultural treatment will result in trade-offs in either volume or in diameter (or quality). The economic feasibility of any treatment regime is dependent on the relative values of costs and benefits and on the timing of their occurrence. Stand level optimization of value of various silvicultural treatment regimes should be carried out and the ones that gives the highest net preent value and low sensitivity to rotation ages should be selected for forest level implementation. Sensitivity of net present values to rotation age is very important as forest level harvesting invariably has constraints that prevent the harvesting of the stand at optimum age. The more severe the constraints (e.g. even-flow) the more variations will occur in the age of harvest of stands at the forest level. In these cases a larger window of feasible rotation ages should be identified for forest level harvesting.  89 In this research, due to lack of resources, optimization at the stand level was not undertaken.  The selection of rotation age was based on the culmination age of MAI of  an untreated stand.  Based on the stand level economic analysis, the following  silvicultural regimes that showed the highest returns were selected for inclusion in the three silvicultural intensities at the forest level. •  Basic intensity: natural regeneration of all harvested stands  •  Medium intensity: -artificial regeneration of six stand groups belonging to Douglas-fir (SI=12 & 19), western red cedar (SI=13 & 21) and Engelmann spruce (SI=1O & 18). -PCT to 800 sph for Douglas-fir (SI=19), western red cedar (SI=21) and Engelmann spruce (SF1 8)  •  High intensity -artificial regeneration of six stand groups belonging to Douglas-fir (SI=12 & 19), western redcedar (SI= 13 & 21) and Engelmann spruce (S1 10 & 18) -two applications of fertilizer to Douglas-fir (SI=19) -commercial thinning to 1/3 rd volume for Engelmann spruce (SI=1 8) -commercial thinning to 1/2 volume for Douglas-fir (SI=19) and western red cedar (SI=21) -two levels of pruning for Douglas-fir (SI=19), western red cedar (SI=21) and Engelmarm spruce (SI=1 8)  90  7 FOREST LEVEL ANALYSIS  7.1 CURRENT MANAGEMENT PRACTICES Simulation runs on Revelstoke 1 shows that the current management practices on a 120 year planning horizon, with all resource emphasis rules in place, will yield an even3 per year. flow timber supply of 14,000 m  This timber supply could be increased to a  3 by relaxing all harvesting constraints imposed by the resource maximum of 35,000 m emphasis rules. In economic terms, the timber rent of 18 million dollars (at 3% discount rate) that is generated with current management practices could be increased by about three and a half times to 64.2 million dollars under unconstrained timber production. Current management practice is a mix of rules applied to different resource emphasis areas.  7.1.1 Impact of resource emphasis rules on timber supply Each of the thirteen resource emphasis rules affect timber supply to different degrees according to adjacency, cover constraints and allowed maximum disturbance rates. Impacts of these rules on timber supply are determined by systematic imposition of different resource emphasis rules on the unconstrained condition of the entire land base. The results of the analysis are illustrated in Figure 22.  The economic timber rent  determined by an economic analysis of the timber supply scenarios under the various resource emphasis rules is illustrated in Figure 23. The opportunity costs of foregone revenue in terms of harvestable timber values of some of the key resource emphasis rules  91 are illustrated in Figure 24. The impact of each rule, in terms of timber supply volume and rent, are discussed under the three resources; timber, wildlife and visual quality.  Current Mgt VPR & WIife(h) VPR & Wflfe(m) 0  2  VPR & WUfe() VPR VM&Wr-’.-’ VM & Wlife(m) VM & Wlife(I) VM Wlife(h) Wlife(m) Wlife(I) limber limber(U) 20  30  40  50  80  70  80  90  160  Percent of unconstrained supply (% Volume)  Figure 22 Impact of resource emphasis rules on timber supply. (Refer to Table 1 (page 8)for description ofcodes).  92  Current Mgt VPR & Wlife(h) VPR & WUfe(m) VPR & WUfe(I) VPR VM & Wlfe(h) VM & Wlife(m) VM & Wlife(I) VM Vh)  0  —*  WIife(m) VJ1ife(I) limber llmber(U) -10  10  0  20  30  40  Rent (million  -  50  60  70  $)  Figure 23 Impact of resource emphasis rules on rent. (Refer to Table 1 (page 8)for descrztion ofcodes).  VPR 0 I  (0 0  VM  E 0  Wlife(m) 0  U, 0  limber r:: 0  V  -  10  20  30  —  40  /  ,—  60  50  Opportunity cost (million  .V 70  80  $)  Figure 24 Opportunity cost in terms of timber values of selected resource emphasis rules when applied individually to the entire land base. (Refer to Table 1 (page 8) for description ofcodes).  93 7.1.1.1 Timber emphasis Rule 2 emphasizes the production of timber with a minimum flow of other values for wildlife habitat and visual quality. This is achieved through adjacency and other constraints relating to cover and maximum disturbance rates. The timber emphasis rule causes a sharp decline to 68 percent from the unconstrained case. When each of the constraints (adjacency, cover constraints and disturbance rates) that are built into the timber emphasis rule were examined independently, it was found that it is the adjacency constraint that restricts the timber supply to this level (Figure 25). In economic terms, the loss in timber rent due to this rule is equivalent to 24.8 million dollars.  Adjacency  .  1  Disturbance  0 C.)  Co’er  U  10  20  30  40  50  60  70  80  90  100  Percent of unconstrained supply (%Volume)  Figure 25 Impact of adjacency, disturbance and cover constraints of timber emphasis rule on timber supply.  94  7.1.1.2 Wildlife emphasis A total of nine rules (3, 4, 5, 7, 8, 9, 11, 12 & 13) emphasize the maintenance or enhancement of wildlife quality. Essentially these rules have a 40 percent maximum disturbance rate and some forest cover requirements. The first three rules (3, 4 and 5) that are an enhancement of the timber emphasis rule, bring about an additional reduction in timber supply volume of about 6 to 10 percent of timber emphasis rule.  This is  equivalent to a further reduction in timber rent of 4.3 to 10.6 million dollars relative to the timber emphasis rule. The opportunity cost for the rule which emphasizes medium quality wildlife is 29.4 million dollars. The second and third set of three rules (7, 8 & 9 and 11, 12 & 13) are an enhancement of wildlife quality along with visual quality modification and visual quality partial retention, respectively. In these cases, the maximum disturbance rate is lower than that for wildlife. As such, when wildlife emphasizing rules are associated with visual quality constraints, their impacts both in terms of timber supply and timber rent are not apparent.  7.1.1.3 Visual quality A total of eight rules (6, 7, 8, 9, 10, 11, 12, & 13) emphasize the maintenance or enhancement of visual quality. The first set of four rules (6, 7, 8 & 9) involve visual quality modification where the allowed maximum disturbance rate is 25 percent.  These  95 rules reduce the timber supply to 49 percent of the unconstrained supply.  The  opportunity cost of visual quality modification is 39.1 million dollars. The second set of four rules (10,11,12 & 13) involve visual quality “partial retention” where the allowed maximum disturbance rate is 10 percent.  These rules  reduce the timber supply to 7% of the base case. This gives a negative rent of about 6.4 million dollars because the small revenue generated towards the end of the planning horizon is heavily discounted, and is not sufficient to cover costs generated in the earlier periods. In other words, the opportunity cost in terms of timber rent forgone for visual quality with partial retention is approximately 70 million dollars. 7.2 TIMBER SUPPLY UNDER ALTERNATIVE LAND USE SYSTEMS  7.2.1 Alternative land use systems Two types of land use systems were identified. One is the integrated use system and the other is the single use system.  7.2.1.1 Integrated use system (IIJ) In this case, the whole of Revelstoke 1 (covering 17575 ha) is considered as an integral unit of production where timber and other uses are produced on every hectare. The distribution of resource emphasis areas under integrated use system is illustrated in Appendix Figure 15. Current management practices on a planning horizon of 120 and 3 and 11,800 m 3 per year, respectively. 240 years produce a timber supply of 14,300 m The 240 year value is lower because the high volume, old-growth stands have been liquidated (fall down effect).  96 7.2.1.2 Single use system (SU) This system includes a zone specializing in the production of timber rather than one integral unit of production.  The balance of the area is devoted to the production of  multiple uses other than timber (i.e.: no timber production in these zones).  The timber  zone in this case is the minimum area which, under unconstrained conditions, could produce periodic timber volumes equivalent to that produced under current management practices from the whole of Revelstoke 1. The single use area was identified by repeated iterations with ATLAS simulations to select contiguous zones that satisfied the volume 3 on 120 year PH and 11,800 m 3 on 240 year PH). This method has targets (14,000 m been discussed in detail elsewhere (Sahajananthan 1994). The single use, area or timber production zone, covers 5362 ha which is about 31% of the gross area and 46% of the net area of Revelstoke 1. The distribution of timber zone and integrated use zone under the single use system is illustrated in Appendix Figure 16.  7.2.2 Timber supply under alternative land use systems Harvest scenarios under both integrated use and single use systems were modeled using ATLAS. Evenflow volumes on a 120 year planning horizon for the integrated use 3 3 and 14,300 m (IU) system and single use (SU) system were determined to be 14,000 m per year respectively. For a 240 year planning horizon, the evenflo volume for the IU /yr for the SU system. On a 120 year planning 3 system is 11,800 m /year and 11,200 m 3 period, single use evenflow is marginally higher than that of integrated use because of possible aggregation of more harvest cut-blocks (i.e., individual yield units within zones)  97 than is required to produce an equivalent volume when entire contiguous zones were aggregated. This slight increase in harvest flow in the single use system does not affect either the methodology or outcome of this research. Composition of harvest area by forest types in the IU system and SU system over a 120 year planning horizon is illustrated in Figures 26 and 27.  The following  observations can be made. i)  The second-growth requires a larger area to produce an equivalent volume to that produced by old-growth.  This is because of: i) the smaller volume harvested at  rotation age from the second-growth compared to old-growth at an older age, and ii) an 11 percent reduction in productive area occupied by roads and skid trails. ii)  Of the total area harvested during the planning horizon, the second-growth stands constitutes 24 % in the single use system and only 7% in the integrated system. This is because more old-growth is available in the IU system. In the SU system, there is a need to switch more quickly to second-growth.  iii)  In both systems, the second-growth is harvested from decade nine onwards. This means that modified forests that were established by converting old-growth and second-growth can only be harvested beyond the ninth period, unless the harvesting priority system is changed.  As such, any investments  in the form of intensive  silviculture on the modified forests will not likely be realized during a 120 year planning horizon. Due to the third observation above, it was decided to extend my analysis to a 240 year planning horizon instead of 120 years.  Rent generated on a 240 year planning  98 horizon is slightly lower than that with 120 year planning horizon because there is a reduction in the evenflow harvest caused by relatively low volumes in the regenerated stands and also due to discounting of net benefit for the period 121  -  240.  The rent at a 2% discount rate generated under JU on a 120 year planning period amounts to $17.9 million while under SU it amounts to $17.1 million. The rent does not appear to be commensurate with the slightly higher evenflow volume seen under single use. This is because of higher harvest system costs associated with the areas selected for single use timber production (which was not intentional).  700 600 500 400  •Sec-growth  300  • OIdgrowth  200 100  10 year periods  Figure 26 Integrated use system showing composition of harvest area by forest types over the 120 year planning horizon.  99  500 400J  • Sec-growth • Oldgrowth  300] .  ,  200  10 year periods  Figure 27 Single use system showing composition of harvest area by forest types over the 120 year planning horizon.  7.3 INTENSIVE TIMBER MANAGEMENT ON A 240 YEAR PLANNING HORIZON  Simulations with the three levels of intensity (basic, medium and high) were completed for both land use systems. The results are discussed under the categories of: i) maximum evenflow volume, ii) rent, iii) delivered wood costs, and iv) environmental indicators.  7.3.1 Economic parameters  7.3.1.1 Maximum evenflow volume The maximum evenflow volume that could be obtained under varying intensity levels of timber management was estimated. The results are illustrated in Figure 28.  100  16 14 12 1o E 0 o8  .Iu .s U  Basic  Medium  High  Management Intensity  Figure 28 Impact of intensive timber management with integrated (IU) and single use (SU) systems on maximum evenflow volume on a 240 year planning horizon. (Refer to Table 1 (page 8)for descrzption ofcodes).  With the basic level of management, the IU system yields a higher maximum /yr) than the SU system (1 1,200m 3 even-flow(11,800m /yr). This is because in a SU 3 system, the slow rate of growth of the second-growth coupled with alimited availability of old-growth constrains the harvest level. The JU system, on the other hand, places many constraints on timber production, so the harvest level throughout Revelstoke 1 is low. This results in slow liquidation of the old-growth, relative to the SU system, hence a slightly higher harvest.  With medium and high intensity levels of management, the  /yr). 3 situation is reversed where the single use system yields a higher evenflow (15,700 m In these cases, the faster growth rates exhibited by the modified forest overcomes the limitations of limited area and thus leads to a higher maximum flow (i.e., SU can capture these increases while IU cannot).  __i.Iu ___  101 7.3.1.2 Timber rent Timber rent was calculated at a 2 percent discount rate, using real price increase assumptions, for both land use systems at the three intensity levels on a 240 year planning horizon. The results are illustrated in Figure 29.  Results show that the SU system  generates a higher rent than the IU system at all levels of management intensity. At the basic level, rent from the SU system is 118% of that from the IU system. This is because of the lower administrative costs in the SU system relative to the IU system resulting from the concentration of the harvest in the timber zone.  At the’ medium and high  intensity levels the rent is as high as 216% of integrated use at the basic level.  This  results from the increased value and levels of harvest during the second half of the planning horizon, and from the reduction in length of roads opened and maintained during each planning period.  30 25 44  c20. 15-  •su  —1  ‘10  5 0 Medium management intensity  Figure 29 Impact of intensive timber management with integrated use (IU) and single use (SU) systems on rent (2 % discount rate on a 240 year planning horizon). (Refer to Table 1 (page 8)for description ofcodes),  102 7.3.1.3 Sensitivity of rent to discount rates Sensitivity of rent to discount rates was tested by determining the rents at 0% and 4% discount rates.  Results of this sensitivity are illustrated in Figures 30 and 31. The  pattern of rent at both discount rates is consistent with that seen at 2% where the SU system rent is higher than that from the IU system.  But the absolute rent is highly  sensitive to changes in the discount rate. At 0%, SU rent at medium and high intensities becomes as high as 424% and 444% respectively, relative to IU at basic intensity. At 4% the, difference between systems is not so sharp. Rent from medium and high intensity management has fallen because the beneficial effect resulting from high value timber that is harvested in the second half of the planning horizon is heavily disco!lnted.  450 400 350 c 0  300 250  .Iu [su  200 150  &  100 50 0 Basic  Medium  High  Management intensity  Figure 30 Impact of intensive timber management with integrated use (IU) and single use (SIT) systems on rent (0 % discount rate, 240 year planning horizon).  103  7 6  0  E4 E  •lu • sul  0i Medium Management intensity  Figure 31 Impact of intensive timber management with integrated use (IU) and single use (SU) systems on rent (4% discount rate, 240 year planning horizon). Refer to (Table 1 (page 8) for description ofcodes).  7.3.1.4 Sensitivity of rent to changes in price of logs Sensitivity of the rent to a 12% change in price of all logs shows that all IU systems are more sensitive to prices than the SU systems. This is illustrated in Figure 32. This happens because of the higher percent of old-growth and second-growth that gets harvested in the IU system. The prices of timber from old-growth and naturally grown second-growth are relatively higher priced than the timber from modified forests. Therefore, when a higher proportion of old-growth gets harvested in a scenario, the change due to price changes is higher. This is seen at the basic intensity levels in both IU and SU systems. The reduction in rent associated with a change in prices is more at the medium intensity because the harvest contains higher value timber than any other scenario. At low prices the total premium on high value logs gets reduced, and there is a  104 proportionately greater reduction in rent.  At high intensities, low value thinnings  contribute little towards rent and therefore this scenario is less sensitive to changes in price. 160 140 120  LEE  _-  _•_  12% low Current  __i2% high  2of  lu_i  IU_2  IU_3  SU_i  SU_2  SU_3  Land use systems and their management intenstiles  Figure 32 Percent change in rent to increase and decrease in log prices by 12%. (Refer to Table 7 (page 68) for description ofcodes).  7.3.1.5 Delivered wood costs Several observations can be made about the delivered wood costs in the two land use systems. z) First, delivered wood costs in the SU system are slightly higher than that for the integrated use system. The components of the delivered wood cost (harvest system, road construction and maintenance, and the hauling costs) are illustrated for medium intensity timber management for both land use systems in Figure 33.  Harvest  105 system costs under single use constitute 72 % of delivered wood cost as opposed to 69 % in the integrated use. This is caused by the more difficult terrain conditions in the single use area. Had areas with identical terrain conditions been selected, harvest system costs in both systems would have been equal. Road costs and hauling costs are lower in the case of single use. This is understandable as there are fewer active roads and shorter hauling distances. Also, new roads are being constructed in a more limited area. ii)  Second, road costs constitute only about 5% of the delivered wood costs.  The road  costs are generally front loaded in both systems and decrease gradually towards the end of the planning horizon. Figure 34 illustrates the pattern of periodic road costs over time for single and integrated use with medium intensity timber management. iii)  Third, delivered wood costs in both systems decline over time. This trend is caused by the declining road costs. 35 30 25 20  0 Hsys_C  15  [oad_C  C,  10 5  0 SU_2  IU_2 Land use system  Figure 33 Delivered wood costs in integrated use (IU) and single use (SU) systems with medium intensity management showing its components viz., hauling cost,  106 harvest system (hsys) cost, and road construction and maintenance costs (road_C). (Refer to Table 1 (page 8)for description ofcodes).  1.20  i.oo  i :: 0.40  1  2  3  4  5  6  78910  11  12  20 yr planning periods  Figure 34 The IU and the SU systems at medium management intensity showing periodic cost of road construction and maintenance on a 240 year planning horizon. (Refer to Table 1 (page 8)for description ofcodes).  Some general observations can also be made on the possible behavior of the components of the DWC in the SU and the IU systems. Since the timber zones in the SU systems will be selected mostly in less difficult areas, its harvest system costs are likely to be very much lower than that of the IU system. In this study, though the timber zone was selected from productive sites, less attention was paid to terrain conditions and their associated costs. This is why the harvesting system costs in the SU are higher than that of the IU. With respect to the road costs, the SU system is likely to cost less than that of the IU because of the smaller area covered. Expenditure on roads during the early part of the planning horizon in the IU system will be very much more than in the SU system as the harvesting areas in the IU system will be more dispersed. As for the SU system, road  107 construction can be spread over most of the planning horizon depending on the progress of harvesting and silvicultural activities. Further, with the present day requirements for deactivation of roads (when not actively used for more than a specific period of time) the IU system road costs will increase several fold. As for the SU system, the limited length of roads will be maintained throughout the year and it will have less deactivation and activation costs. As for hauling cost, it is difficult to predict the hauling distance for the two systems. If, however, it is assumed that both the SU and the IU systems are placed at equal distance from the mill site, then the SU system costs will be slightly lower than that of the IU system due to the concentration of harvesting within the operable harvesting area.  7.3.2 Environmental parameters This section summarizes landscape pattern responses to IU and SU management, in terms of landscape indices calculated by SIMFOR and ATLAS, and discusses their biological significance.  7.3.2.1. Seral Stages Results of the simulation runs on the SU and the JU systems at the basic timber management intensity showing the distribution of seral stages for all of Revelstoke 1 over a 240 year planning horizon are illustrated in Figures 35 and 36.  These Figures show  that, in both cases, the very old-growth (seral stage 5) increases from 29% to nearly 70%, while the regeneration, pole and mature seral stages each stabilize at 10 % of the area  108  around period 7. Both the IU and the SU systems appear quite similar as nearly 34% of the Revelstoke 1 is made up of reserves that cannot be harvested. The pattern of seral stage distribution is similar for SU and IU systems at medium  and high timber intensities. At medium timber management intensity, the regeneration seral stage drops to about 7% of the total area near the end of the planning horizon in both SU and IU systems. This is because of the larger volume (resulting from PCT) harvested from a smaller area. At high intensity, the area of the regeneration seral stage is slightly higher (12%) than expected (7% to 8%) because the model does not distinguish between commercial thinning and clear felling operations. The evenflow volume management strategy adopted in this study results in the harvest of approximately equal areas. As such, no significant differences in seral stage distribution can be observed between SU and IU systems.  Since evenflow volume  management prevents the harvest of available higher volumes during the earlier part of the planning horizon, this leads to a higher proportion of old seral stages. Had an areabased management strategy been adopted, the accumulation of old-growth could have been reduced. The simulation with area-based management would also have shown the time it takes for the forest in both systems to normalize. Natural calamities such as wildfire and insect attacks also affect the periodic distribution of seral stages in both systems. However, since it is assumed that both the IU and the SU systems will be affected in the same way, the natural calamities do not affect the outcome of the comparison.  109  (5 (5 0 >  •eryold  (5  D old O mature •pole • regen  E (3 (0 Cu (0 (5 (5 (0 (5 (0  0  1  2  3  4  5  6  7  8  9  10  11  12  20 year period  Figure 35 Single Use showing the distribution of seral stages for basic intensity management over a 240 year planning horizon. (Refer to Table 1 (page 8) for description ofcodes)  (5 (5  •.eryoId  0 >  H sold mature O  (5  E  • pole .regen  U (5 (5  en (5  (a (5 0  en  20 year period  Figure 36 Integrated Use showing the distribution of seral stages for basic intensity management over a 240 year planning horizon. (Refer to table 1 (page 8) for description ofcodes).  110 7.3.2.2 Ecosystems represented in seral stages The results of the simulations on the SU and the IU systems at basic timber management intensity showing ecosystems represented in the very old-growth (seral stage 5) stage are illustrated in Figures 37 and 38. The overall pattern appears similar in that the percentage of very old ICHmw stands increases substantially in both the SU and IU systems. But distinct differences in the percentages of ecosystems represented in each system can be observed. Ecosystems represented by very old stands in each land use system as a percentage of all ecosystem areas at the start and at the end of the 240 year planning horizon are given in Table 14.  Table 14 Ecosystem types represented in very old-growth (>240 years) as percent area of total land base at the start and end of the 240 year planning horizon. (Refer to Table 1 (page 8)for descrztion ofcodes) Ecosystem type  Start ofPH (% area)  End ofPH (% area)  sU&IU  su  lu  ICHmw  8  46  36  ICHwk  15  10  17  ESSF  5  13  14  From Table 14, it can be seen that the percentage representation of all three ecosystem types, except for ICHwk in SU, increase at the end of the planning horizon. This is because of the accumulation of old-growth during the planning horizon as discussed in Section 7.3.2.1. Since more area has been harvested in the ICHwk type in  111 the SU system, this type shows a decrease. As stated earlier, the ecosystem data used in this analysis did not have details related to biogeoclimatic subzones and variants. Had these been available, more information on the pattern and distribution of ecosystem types in each of the seral stages over the planning horizon would have been possible. These results depend on the original distribution of ecosystem types both within and outside the timber zone. If most of the timber zone is dominated by one or two ecosystem types, then it is obvious that those old-growth forests will be harvested under the SU system. In multiple use management strategies, it is important to maintain a certain percentage of the total area as a network of reserves comprised of each ecosystem type. The percentage to be maintained should be based on the natural distribution and on critical habitats. This type of analysis will help to determine whether all ecosystem types  are adequately represented in the system.  70 60 50 E  40  ESSF 0  30  • ICHwk ICHmw,  00  201 10 0 0  1  2  3  4  5  6  7  8  9  1011  12  20 year period  Figure 37 Single Use showing the distribution of ecosystem types within very oldgrowth seral stage (> 240 years) for basic intensity management on a 240 year planning horizon. (Refer to Table 1 (page 8) for description ofcodes)  112  ho DESSF  40  &CHwk  30  .ICHmw  20 10 0 0  0  1  2  3  4  5  6  7  8  9  10  11  12  20 year period  Figure 38 Integrated Use showing the distribution of ecosystem types within very old-growth seral stage (>240 years) for basic intensity management on a 240 year planning horizon. (Refer to Table 1 (page 8)for description ofcodes).  7.3.2.3 Edge habitats Edge habitat is a band of interface between old-growth and regeneration or pole stages (defined in Sections 3.2 and 5.4.3). The percentage area of old-growth (>120 years) that has edge habitats under the SU and the IU systems with basic, medium and high intensities of management are illustrated in Figures 39, 40 and 41, respectively. Edge influence on old-growth depends on several factors. Some of the key factors are: i) size and shape of cut blocks; and ii) size and distribution of patches of remnant oldgrowth in the management unit. Some obvious correlations to certain notable features of the landscape can be observed and are discussed in this section.  113  20 year periods  Figure 39 The SU and the IU systems showing area of edge habitat (as percent area of old-growth (>120 years) and very old-growth (>240 years) at basic intensity management on a 240 year planning horizon. (Refer to Table 1 (page 8) for description ofcodes)  16.0  14.0 12.0 10.0 su 8.0  _._Iu  6.0. 4.0. 2.0. 0.0 0  1  2  3  4  5  6  7  8  9  10  11  12  20 year periods  Figure 40 The SU and the IU systems showing the area of edge habitat (as percent area of old-growth (>120 years) and very old-growth (>240 years) for medium intensity management over a 240 year planning horizon. (Refer to Table 1 (page 8) for description ofcodes)  114  j 20 year periods  Figure 41 The SU and the IU systems showing the distribution of edge habitats (as percent area of old-growth (>120 years) and very old-growth (240 years) for high intensity management over a 240 year planning horizon. (Refer to Table 1 (page 8) for description ofcodes)  The average area of edge habitat over the planning horizon as a percentage of the old-growth for basic, medium and high intensities area is given in Table 15. Figures 39 through 41 indicate that, currently, 11% of the area constitutes edge habitat. In basic, medium and high silvicultural intensities it is seen that the edge habitat (expressed as an average percentage of the old-growth for the whole planning horizon) in the IU system (12%) is more than twice that of the SU system (5%).  One reason for this is the  cessation of felling in nearly 60% of the area which results in closing of the edge habitats. In the SU system, edges stabilize at around 4%, possibly after closing up of the canopy in the non-timber zone.  115 Table 15 Average area of edge habitat as percent area of old-growth habitats, and average area of regeneration edge as percent of regeneration area over the 240 year planning horizon. (Refer to table 1 (page 8) for description ofcodes). Average old-growth edge  Average regeneration edge  (as % old-growth area)  (as % regeneration area)  su  lu  su  lu  Basic  5  10  10  44  Medium  5  11  11  42  High  6  14  14  45  Intensity  At medium intensity the smaller area harvested in the second half of the planning horizon appears to influence the percentages. As such, both the IU and the SU systems show a lower than average edge percent near the tail end of the planning horizon. At high intensity, the average is higher than those at basic and medium intensities because of the commercial thinning. In this research, commercial thinning is assumed to disrupt the interior habitat since one third to one half of the available volume is removed at these thinnings. Generally, the percent area of edge habitat in an undisturbed natural forest is likely to be very much lower than that in a managed forest.  Edge habitats are being  portrayed as both good for game species (Dasmann 1964; Thomas et al. 1979; Lovejoy et al. 1990) and bad particularly for non-game species and interior dependent wildlife species (Wilcove 1985; Wilcove et al. 1986; Noss 1991) by many scientists. But there is a fine distinction between the edge habitat formed by natural disturbance and by forest management practices. Edges created under natural conditions are not sharp and would  116 have reached a stage of equilibrium with the interior and exterior of the patch thus providing  special habitats for some animals.  Whereas the edge created by forest  practices are generally sharp and most would not have reached the stage of equilibrium with the interior and exterior of the patch due to frequent disturbances caused by harvesting. Thus the habitat found in these edges are likely to differ from those created under natural conditions.  7.3.2.4 Influence of old-growth edge on regeneration The microclimate within the regeneration area is always affected by the adjoining old-growth stands. This altered microclimate within the regeneration area influences the seedling establishment and competitive interactions between individual plants results in changes in forest structure and composition (Saunders et al. 1991; Chen et al. 1992). For example, growth rates of shade-intolerant species such as Douglas-fir may be reduced in a wide band along the south side of a plantations due to shading from adjoining forests (Hansen et al. 1993). The depth of edge influence on regeneration area estimated in this research is given in Section 5.4.3. The area of regeneration affected by edge will depend on the size of cut blocks, and the relative area occupied by the regeneration seral stage at a particular time. The results are illustrated in Figures 42, 43 and 44. The results indicate that like the old-growth edges, here too the regeneration area affected by old-growth edge in the IU area is more than twice that of the SU area. Average regeneration area affected for basic, medium and high intensity silvicultural  117 regimes are given Table 15.  The results also indicate that 42% to 45% of the  regeneration area is affected by the edge effects of the old-growth. At high intensity of timber management the area affected is the highest (45%).  This is because of the  standard adjacency requirement in commercial thinnings which spreads the thinnings over much of the landscape. The percent of regeneration edge affected (e.g., 42% 45% -  in IU) appears higher than edge habitat (e.g., 10%  -  14% in IU) in old-growth because the  area of regeneration is relatively small when compared to the area of old-growth in any specific period. If forest practices rely only on natural regeneration for restocking the forest, it is likely that species established in the center of the regeneration area will be slightly different from those at the periphery, due to differences in seed dispersal effectiveness and changes in the microclimate. Consequently, this may require different silvicultural approaches to management. Shade and other microclimatic effects due to old-growth adjoining the regeneration areas will affect the establishment and growth of both natural regeneration and planted seedlings in the edge zones, and will thus reduce overall secondgrowth stand productivity for some time (Bradshaw 1992).  Since nearly 40% of the  regeneration is affected in this way, it also possible that artificial regeneration and other related intensive management efforts may encounter difficulties.  118  50.0 (U (U  40.0 (U (3  30.0  (U C 0  20.0  (U (U C (U  10.0 0.0 0  1  2  3  4  5  6  7  8  9  10  11  12  20 year periods  Figure 42 The SU and the IU systems showing the area of regeneration affected (as percent of the regeneration area) by old-growth and very old-growth edges for basic intensity management on 240 year planning horizon. (Refer to Table 1 (page 8) for description ofcodes).  50.0 (U (U  40.0  (U  30.0  I  _._su —‘U  20.0 10.0 0.0 0  1  2  3  4  5  6  7  8  9  10  11  12  20 year periods  Figure 43 The SU and the IU systems showing the area of regeneration affected (as percent of the regeneration area) by old-growth and very old-growth edges for medium intensity management on 240 year planning horizon. (Refer to Table 1 (page 8) for descrzption of codes).  ________ _____  119  60.0 50.0 40.0  30.0  0  1  2  3456789  10  11  12  20 year periods  Figure 44 The SU and the IU systems showing the area of regeneration affected (as percent of the regeneration area) by old-growth and very old-growth edges for high intensity management on 240 year planning horizon. (Refer to Table 1 (page 8) for description ofcodes).  7.3.2.5 Patch sizes Various patch sizes and their distribution among five seral stages were examined. Different seral stages can form habitat for various animals. There is much concern about possible endangering critical habitats found in seral stage 5 (>240 years). The results of the analysis showing the response of patch sizes in seral stage 5 to forestry practices under the SU and the IU systems at basic, medium and high intensities are illustrated in Figures 45 through 50. The average area of each patch size (as percent of total land base) under the SU and IU systems for the three intensities of management over the planning horizon are given in Table 16. The following general observations can be made on these results.  120 z)  At the start of the planning horizon the management unit does not have any patches larger than 1000 ha in size. It is dominated by a patch size 3 (501 -1,000 ha) covering nearly 48 % of the total area.  ii)  The largest patch size (>1000 ha, patch size 4) appears at different periods during the planning horizon and comes to dominate the landscape at the end of the planning horizon.  iii)  At all intensities of management, the smaller patch sizes (patch size 1 and patch size 2) cover a larger area in the IU system than in the SU system.  iv) At all intensities of management, the largest patch size (patch size 4) cover a larger area in the SU system than in the IU system (Table 16).  Table 16 Average area covered by the four types of patches in the very old-growth seral stage (indicated as percent area of total land base) at the end of 240 year planning horizon. (refer to table 1 (page 8)for descrzption ofcodes). Patch size (ha)  Code  Medium (% area oftotal land base)  su  iu  (ha) 6  (ha) 7  High (% area of total land base) su lu (ha) (ha) 6 8  0-100  pat_i  Basic (% area of total land base) su lu (ha) (ha) 7 6  101-500  pat2  ii  15  12  17  12  16  501- 1000  pat_3  5  5  4  5  4  6  >1000  pat_4  22  16  20  12  20  11  121 At basic intensity, both in the SU and in the IU, patch size 4 (>1000 ha) is formed from the first period onwards and continues to increase until the end of the planning horizon, when it covers 71% of the area in the SU system and 64% of the area in the IU system. The reason for the larger area under SU is due to the moratorium on felling in areas other than the timber zone. As for the IU system, the high percent of patch size 4 is caused by the evenflow management strategy which effectively prevents higher harvest levels during some periods even when areas are otherwise available for harvest. At medium intensity in the SU system, there is a steady increase in area covered by patch size 4 from period 1 until it covers nearly 70% of the area in period 12. Patch size 3 disappears from the scene either by being aggregated into larger patches or by fragmenting to smaller patches in periods 5, 9, 10, and 11. In the IU system, on the other hand, the large patch sizes (patch 4) appear only from period 7 onwards and steadily increase to cover 64% of the land area in period 12. The patch size 2 (501-1000 ha) nearly doubles its area from period 2 to the end of the planning horizon by possibly fragmenting patch size 3. Therefore, the total area covered by patch size 3 decreases and drops to as low as 12% for the whole area during periods 2 through 5. During these periods, the largest patch size existing is patch size 3. These landscape structures may have severe biological consequences. For example, they are likely to affect wildlife that have a minimum home range requirement. If this is true, then it can be predicted that intense management cannot be practiced in IU systems that show this type of landscape pattern response.  122 At high intensities, the SU system again shows a steady increase in patch size 4 area from period 1 to the end of the planning horizon, where it reaches 71% of the total area. From period 9 to end of the planning horizon, patch size 3 disappears from the scene. In the IU system at high intensity, patch size 4 appears only in period 2 and it disappears in periods 4, 5 and 6, thus creating a gap. In period 5, when there is not even a single patch size 4, the average area covered by patch size 3 is only 988 ha (16% of the area). From periods 2 through 11 the area under patch size 2 is nearly one and a half times that found in the SU system. In summary, it can be stated that at high intensity, the IU system has not only a higher proportion of smaller patch sizes and a small proportion of large patch sizes, but it also has gaps in the larger patch sizes. These are likely to have serious biological consequences some of which are discussed Chapter 8.  I  pat_4 0  8000 .  o pat_3  6000  • pat_2 • pat_i  4000  20 year periods  Figure 45 The SU system showing the distribution of patch sizes (as area in ha) in very old-growth (>240 years) for basic intensity management over a 240 year planning horizon. (Refer to Table 1 (page 8)for descr4#ion ofcodes).  123  12000 10000 pat4 0  8000  pat3 0 .  ! N  6000  • pat2 • patl  4000  0  .5  2000 0 0  1  2  3  4  5  6  7  8  9  10  11  12  20 year periods  Figure 46 The IU system showing the distribution of patch sizes (as area in ha) in very old-growth (>240 years) for basic intensity management over a 240 year planning horizon. (Refer to Table 1 (page 8)for description ofcodes).  12000 10000 8000  pat_4  -  pat_3 0 6000 4000  • pat_2 •pat_1  -  -  N 0  2000 (V Q.  -  0 0  1  2  3  4  5  6  7  8  9  10  11  12  20 year periods  Figure 47 The SU system showing the distribution of patch sizes (as area in ha) in very old-growth (>240 years) for medium intensity management over a 240 year planning horizon. (Refer to Table 1 (page 8) for descr4’tion ofcodes).  124  12000 ‘C  10000  0  ‘a o..  pat_4  8000  pat_3 0 6000  • pat_2 • pat_I  4000  N 0,  2000  2 0.  020 year periods  Figure 48 The IU system showing the distribution of patch sizes’ (as area in ha) in very old-growth (>240 years) for medium intensity management over a 240 year planning horizon. (Refer to Table 1 (page 8)for description ofcodes),  12000 10000  I N  -  pat_4’  8000 6000  O pat_3 • pat_2  4000  • pat_i  en ‘C  2 0.  2000 0 0  1  2  3  4  5  6  7  8  9  10  11  12  20 year periods  Figure 49 The SU system showing the distribution of patch sizes (as area in ha) in very old-growth (>240 years) for high intensity management over a 240 year planning horizon. (Refer to Table 1 (page 8) for description ofcodes).  125  12000 10000 pat_4  8000  3 Qpat_ .  6000  • pat_2 • pat_I  4000 N  (a,  .5  2000  9  10  11  12  20 year periods  Figure 50 The IU system showing the distribution of patch sizes (as area in ha) in very old-growth (>240 years) for high intensity management over a 240 year planning horizon. (Refer to Table 1 for description ofcodes).  7.3.2.6 Harvest pattern in old-growth An analysis was done to examine whether all old-growth stands (>120 years) available at the start of the planning horizon would be harvested by the end of the planning horizon in the two systems.  The results illustrated in Figure 51, are very  interesting. Under integrated management at varying intensities of management, 50% to 53% of the original old-growth remains unharvested. Under single use management, a higher percentage, ranging from 62% to 63%, remains unharvested. Even within the timber production zone, approximately 35% of the old-growth remains unharvested because more accessible, modified forest crops rapidly become available for harvesting. If short term harvests had been maximized and subsequently allowed to decline in the long term (rather than even-flow), there would be no old-growth in the single use area. In  126 hindsight, a declining harvest flow policy would have been a better choice for this study. The evenflow policy is too rigidly fixed to the long term, steady state of the forest. The declining flow policy would have pushed both systems to their respective limits, and likely have produced greater differences than observed here. This is especially important for short term timber supplies  -  an issue of great concern to British Columbians.  Implications of altering harvest flow policy are discussed in Chapter 8.  70 60 C  50 40  2  • % OG remaining  30 20  0  10 0 SU_1  SU_2  SU_3  Ui  IU_2  IU 3  Land use system and its management intensity  Figure 51 Percent of old-growth (OG) retained with integrated use (IU) and single use (SU) systems at the end of the 240 year planning horizon (all of Revelstoke 1). (Refer to Table 7 (page 68) for description ofcodes).  7.3.2.7 Density of roads The density of roads in the SU and IU systems was estimated in terms of the average length of roads maintained during a planning period and on the total length of roads constructed to the planning horizon. Results (Figure 52) show that at all three  127 levels of intensity, the length of roads maintained and constructed in single use is less than (65% to 68%) that of integrated use. The SU system requires a smaller area to access than the IU system. The length of roads constructed and maintained is higher at high intensity levels because of the thinnings. At medium intensity levels, fewer roads are required than at basic levels because a higher volume per hectare is harvested. This is mostly due to evenflow policy, otherwise it is likely to be the same across all SU systems.  250 200 E  • Rd open/pd a Rd constl24oyrs  150 i00_ 0  50  -  0 U_i  lU_2  lU_3  SU_1  SU_2  SU_3  Management intensity  Figure 52 Road density with integrated use (IU) and single use (SU) systems at basic, medium and high management intensities showing average length of roads maintained per period, and constructed during the planning horizon. (Refer to Tables 1 (page 8) and 3 (page 50) for description ofcodes).  128 7.4 ENHANCEMENT OF NON-TIMBER VALUES IN THE TIMBER ZONE  7.4.1 Impact on even-flow volume The impact on even-flow volumes of introducing selected wildlife and visual quality rules to unconstrained timber production at the three management intensities is illustrated in Figure 53. Two rules relating to wildlife are tested. The first is the timber emphasis rule that is treated as a basic requirement for maintaining wildlife.  It is  abbreviated as wildlife (basic). The second is rule number 3 or wildlife (high) where the cover constraint is 60% (i.e.,> 60 % of the area must be covered by stands  >  height class  3). Results show that there are minor differences in impact between the basic and high wildlife quality.  A sharp drop in evenflow volumes down to 58% (that is, a 42%  reduction) occurs due to the introduction of adjacency constraints in the timber emphasis rule. The drop is higher on a 240 year planning horizon than on a 120 year planning horizon. Further increases in cover constraints, from 30% with height class 2 to 60% with height class 3, marginally lower the volume by an additional 2%. These results suggest that it may be possible to maintain wildlife with medium or large home ranges and large dispersal distances by modifying the adjacency requirement and by introducing the required cover constraints and disturbance rates, thus avoiding the large sacrifice of timber related to existing adjacency rules. This reduction in volume caused by adjacency could be reduced by resorting to intensive management which at medium and high intensities lower the reduction from 42% to 15% and 30%, respectively.  The lower  volumes at high intensity can be attributed to additional adjacency constraints caused by thinnings. Relaxing adjacency rules for thinnings would alleviate these losses.  129 The introduction of visual quality (modification) lowers volume to 50%, 73% and 57% of the unconstrained (basic intensity) case for basic, medium and high intensities of management, respectively. The lower volume for high intensity management is again due to additional adjacency constraints imposed by thinnings.  Visual quality (partial  retention), reduces the evenflow volumes down to 9% of the unconstrained case. Under these visual rules, intensive timber management practices do not help to increase the timber supply.  This is understandable as the allowed disturbance rate of 10% and  adjacency constraints permit no room for high timber production.  However, there is  some possibility of faster green-up with intensive management, thus resulting in early release of areas locked up in adjacency and cover constraints.  .Basic • Medium High  E a 0 0  E  -I  SU_Wlife(h)  SU_VM&Wlife(b)  SU_VPR&Wlife(b)  Resource emphasis  Figure 53 Impact of wildlife and visual quality emphases on volume from the timber zone(SU). IU system volume is also shown. (Refer to Tables 1 (page 8) and 3 (page 50) for description ofcodes)  130  7.4.2 Impact on rent The impact of selected wildlife and visual quality rules on rent from the unconstrained production of timber is illustrated in Figure 54. Results show that the introduction of the timber emphasis rules lowers the rent to 35% of the unconstrained rent. This reduction can be reversed by intensive timber management. With medium and high intensities, the rent can be increased to 151% and 149%, respectively of the basic level of unconstrained production.  The situation is similar for high wildlife quality  where the reduction at basic levels could be reversed by intensive timber management. With visual quality modification emphasis there is reduction in rent at all intensities of timber management. In the case of visual quality “partial retention”, only negative rents are produced at all intensities of management. The results of the analysis shows that wildlife and visual quality can be accommodated, to some degree, without reductions in current levels of rent. But this is always accompanied by high opportunity costs.  4,  0  E C C  a  Resource emphasis  Figure 54 Impact of wildlife and visual quality emphases on rent from the timber zone (SU). IU system rent is also shown. (Refer to Tables 1 (page 8) and 3 (page 50) for description ofcodes)  131  8 DISCUSSION Managing by resource emphasis rule is an important development in the management of forests for multiple uses in British Columbia. These rules are supposed to ensure a steady flow of various goods and services during the planning period. Some of these rules (adjacency and basic cover constraints) are universally applied while others (some types of wildlife habitats and visual aesthetics) are site-specific. Under integrated use management, the value of timber production is usually residualized, that is, timber production only becomes possible after all resource emphasis rules have been met. Consequently, these rules necessarily have an opportunity cost associated with them in terms of timber values forgone. Simulation of harvesting with variou resource emphasis rules under the SU and the JU systems leads to some of the very interesting results presented in Chapter 7.  These results have far reaching consequences which are  discussed in this section.  8.1 OPPORTUNITY COST OF RESOURCE EMPHASIS AREAS To facilitate the discussion on the opportunity cost of managing the different resource emphasis areas, their equivalent annual flow of opportunity cost in the whole of Revelstoke TSA was determined by extrapolating the results of the study area. This extrapolation was found to be useful as Revelstoke TSA has an area dedicated for timber emphasis which is not present in Revelstoke 1. The extrapolation for wildlife REA was done with the wildlife (medium) rule (Rule 5) and that of visual quality was done with visual quality modification (Rule 6). The results showed that the equivalent annual flows  132 of opportunity costs of REAs for timber emphasis, wildlife(m) and visual quality (modification) were 2 million, 8.8 million and 6.2 million dollars per year, respectively. This result leads to three interesting questions: i) whether the values maintained by the application of these rules are higher or equivalent to the opportunity cost in terms of the value of timber forgone; ii) whether these resource emphasis rules are achieving the desired results over the long period; and iii) whether any other land use system could improve this situation.  Application of adjacency and basic cover constraints in the  resource emphasis area for timber emphasis probably protects wildlife with a small home range at a cost of 2 million dollars per year and at a cost to animals with large home ranges (bear, caribou etc.) Some of the resource emphasis rules prescribing adjacency conditions to disperse the harvest across the landscape are essentially a legacy of the past. In the past, dispersed harvesting was important for several reasons: i) to ensure adequate natural regeneration from seeds from adjoining stands; ii) to maintain the road network for purposes of fire suppression and stand tending; iii) to reduce erosion and sedimentation resulting from harvesting operations; iv) to minimize visual effects; and v) to create edge and early seral habitats favored by game animals (Smith 1986). Now, with improved understanding of the nature of forested landscapes and improved forestry technology, many of these reasons are no longer tenable. Current timber management practices rely more on a few species that are artificially regenerated and possibly genetically improyed, than on natural regeneration. Fire suppression is no longer totally dependent on the road network as interior areas are increasingly accessed by air. It should also be noted that, in many cases,  133 the network of roads found inside the forest is associated with much of the anthropogenic origin of fire (Wallin 1994). Indeed, the high costs of road maintenance in mountainous terrain, and the impacts of overhunting caused by improved access, now favor policies of road deactivation in much of B.C. Since harvesting operations have improved several fold over the earlier methods, erosion and sedimentation rates resulting from these operations are likely to be the same under both the IU system and the SU systems. However, it is likely that more roads in the IU system may cause more sedimentation. As for edge and seral habitats for game animals, the current environmental knowledge suggests that the sustainability of forested ecosystems urgently requires protection of interior habitats, and not the creation of additional edge and seral habitats (Alverson et al. 1994). Further, it is also difficult to have large patch sizes in the future, if you create small patch sizes now. There is also the question of whether the severe restrictions, such as 52% to 60% of the cover requirements, are required across all 64% of the land area at a cost of nearly 8.8 million dollars per year. Much of the wildlife may be protected by making small adjustments in the harvesting pattern and by having set-asides and clumps of snags within the landscape (Sedjo and Bowes 1990). More studies should be undertaken to determine the trade-offs involved between maintaining large cover constraints and setting aside smaller areas of trees and clumps of snags. Resource emphasis areas for visual quality impose the highest opportunity cost. Maintenance of visual quality does not effect ecosystem function.  It is essentially  provided for the consumption of the present generation as there will always be  134 uncertainty about future generations’ demands in this regard. The question that should be asked is, “are we generating visual quality value equivalent to 6.2 million dollars per year?”. If not, ways and means should be sought to reduce this opportunity cost. One way of decreasing this cost is to invest in economically feasible recreation and tourism activities in the region.  8.2 IMPLICATIONS FOR TIMBER SUPPLY  8.2.1 Even-flow volumes Multiple use management of forests in British Columbia is enforced through volume based timber supply management. Under this system, an annual allowable cut (AAC) of timber is determined using biological, social and economic information. This AAC is a subjective decision aimed at maximizing human welfare. simulated even-flow volume calculated at maximum MAI.  This research  There are several  disadvantages in using volume based management particularly when there is a large stock of old-growth and wide variation in site qualities. In this research, four observations can be made about even-flow timber harvests for Revelstoke 1. i) Maximum even-flow timber harvests decrease with longer planning horizons (for /year for 3 example, 14,000 m /year for the 120 year planning horizon, and ‘11,200 m 3 the 240 year planning horizon) of the forest. This is due to the lower volume content of the second-growth compared to old-growth at a later age (that is, the fall down effect). ii) Even-flow timber management leads to a slow liquidation of old-growth in the IU system. This is caused by the difference in available harvestable volumes in various  135 periods. Had the strategy been declining volume flow to LRSY this problem may not have occurred. iii) Even-flow timber management fails to capture the full benefits of intensive timber management. Benefits of intensive management, though reflected in increased levels of even-flow timber supply, are not fully captured due to the ca on the even-flow. This invariably postpones the optimal age window for harvesting. iv) With intensive forest management, the revenue from evenflow harvests will vary widely in different periods, based on the proportion of thinnings and final harvest in each period. This is because thinnings have a low value compared to final harvest. All of these four problems might be solved by resorting to area-based management or by declining volume flow to LRSY. This will allow for early conversion of old-growth and second-growth to managed modified forests. Fluctuations in volume caused by site and species interactions could be adjusted by having some amount of volume control. The effect of this volume control will not be as severe as experienced with pure volume based management. Further, area- based management would be much easier to practice within the timber zone of the SU system than with the IU system, as timber zones are likely to be more homogeneous than the whole area within the IU system.  8.2.2 Rent Rent is affected by the relative values of the products, delivered wood costs (DWC) and administration costs. Since it is assumed that the relative values of products remain constant across both land use systems, they do not affect the outcome for  136 comparative purposes. The higher the DWC the lower will be the rent. The components of the DWC (road costs, harvesting costs and hauling costs) in the SU system can be substantially lowered by suitably locating the timber zones. For example, road costs in the SU system can be very much lowered by locating the timber zones on easy terrain with easy access. This, however, is not possible under the IU system as the roads will have to access all areas. Harvest system costs are usually higher within difficult and sensitive terrain. Under the TU system the resource emphasis rules restrict the harvesting of timber from easy terrain, leading to a necessity to go into more difficult terrain. If timber zones are located on easy terrain, then all harvesting could be done more cheaply. In this research, the timber zone was selected by trial and error on an area which produced the highest volume of timber, and attention was not paid to terrain conditions.  This resulted in  locating the timber zone in difficult terrain where the harvesting costs are slightly higher than average, which has resulted in a slightly higher DWC for the SU system. This shows that a sophisticated technology (incorporating economics and ecology) will have to be developed for selecting timber zones. Hauling cost is also an important component of the DWC. The difference in the hauling cost between the IU and the SU systems depends on where timber zones are placed relative to the whole IU system. The shorter the hauling distance, the lower the costs. Since the timber zones occupy only a portion of the IU land area, the hauling distance for the SU system are shorter than that of the IU system.  137 In this research, it is assumed that adjacency constraints will be applicable to areas subjected to commercial thinnings. The assumption of adjacency is made here because the commercial thinnings remove between 1/3 to 1/2 of the total volume available in the stand at the time of thinning. This operation will create large openings in the canopy and it is very unlikely that these areas will provide sufficient cover or meet winter range requirements of ungulates.  Applicability of the adjacency constraint to commercial  thinning is debatable. It depends on the intensity and the weight of thinnings and it may be possible to have light thinnings (e.g. <25% weight) in areas constrained by adjacency. Adjacency in commercial thinning causes a reduction in the harvest of timber and thus leads to a reduction in rent. In the absence of adjacency constraints, we would expect the harvests to increase. But, since fmal treatment in all stands is clearcut and regeneration with adjacency you still get an adjacency pattern that cannot be easily broken (Wallin et al. 1994). Therefore, there is unlikely to be much difference in rent between commercial thinning systems with and without the adjacency assumption. Administrative cost is an important determinant of the timber rent. It consists of two cost categories: i) management and protection and ii) planning, implementing and monitoring. Administrative costs for the IU system are likely to b much higher than those for the SU system for two reasons: i) a relatively larger area to be managed under the IU system, and ii) higher transactions cost of ensuring a constant flow of non-timber goods and services from every hectare of land. In ensuring the flow of non-timber values, there are additional costs involved in establishing, implementing and monitoring  minimum standards and site-specific guidelines for each resource. For example, once the  138 guidelines have been set to maintain a percent of cover with specific height, there will have to be a specialist monitoring the status of cover types each time the forest is revisited for harvesting. With all these additional expenditures there still remains the question of whether standards are adequate to ensure an adequate flow of other services.  8.3 IMPLICATIONS FOR WILDLIFE The present set of resource emphasis rules in the Revelstoke forest district are designed to protect ungulate wildlife in the forested landscape. Some rules that provide habitat for caribou (with 52% thermal cover) are also expected to adequately provide for the habitat requirements of other species.  The main question that has to be answered is  whether the universal application of these rules is justified. Many of the critical wildlife habitats could be easily accommodated even within the timber zone at very low cost by leaving structures like snags and clumps of old-growth. This would release substantial forests for identifying and protecting various other critical wildlife habitats. Wildlife can be broadly categorized into i) Core (those preferring interior habitats), ii) Edge (those preferring edge habitats) and iii) Neutral (those indifferent to interior and edge habitats) species (Daust 1994). Landscape pattern response analysis showed some interesting results with respect to patch sizes, interior habitats and seral stages represented in the landscape. The size and isolation of patches of wildlife habitat are good indicators of fragmentation and are helpful in analyzing the biological consequences of different forest practices. In a natural environment, the size and isolation of forest patches are generally correlated with groups of environmental variables such as soil type, drainage, slope and  139 disturbance regime (Sharpe et a!. 1987).  But under commercial forestry, these are  dictated by the harvesting pattern and the extent of protection and fire management practiced in the area. Presently, a major concern of the public is the fast rate at which interior natural habitats are disappearing, particularly those associated with natural old-growth areas. Maintaining a mosaic of habitats within old-growth forest depends on the size and isolation of patches of habitats (i.e., the level of fragmentation). Large patches are likely to accommodate a mosaic of habitats. The higher the level of fragmentation, the lower the availability of various habitats. These are particularly important for certain species that require a minimum size or specific arrangement of patches, such as, the spotted owl in the Pacific Northwest (Gutierrez et al. 1985).  The current issue, therefore, is the  identification of suitable patches that will have practical conservation values and how to manage them to retain these values (Saunders 1987). The analysis of the SU and the IU systems shows that maintenance of large patches (>1000 ha) is possible only at basic intensity.  At high irtensities of timber  management, large patches of old-growth become heavily fragmented. This suggests that high intensity timber management may not be compatible with maintaining large patch sizes within the landscape for the IU system as a whole and for the timber zone in the SU system. There is an important difference between patches found in the SU and TV systems. In the SU system, most of the large patches will be maintained pennanently in non-timber zones. A likely management strategy might be to select a network of ideal patch sizes of  140 endemic vegetation, possibly representing each ecosystem subzone. In the case of the IU system, the patches would show temporal variation in their size, distribution and composition.  A patch size 3 (501 ha to 1000 ha) in a particular period may either  aggregate with other patches and grow into a patch size 4 (>1000 ha), or it may be further fragmented and become a patch size 1 or 2. This temporal variation can have serious biological consequences. A patch will include both edge and interior habitat. Under natural conditions they reach a point of equilibrium with time. Natural disturbances, often with long return intervals, keep them in a state of equilibrium. But with integrated forestry practices, the patch sizes in the IU system will change more frequently and all the patches will be in different stages of reaching an equilibrium.  In this equilibrating  sequence, there will be a process of “species relaxation” (Saunders et al. 1991) where some species which can only live in interior habitat are likely to disappear. Meanwhile, there will also be an influx of a suite of new species more adapted to the changing environment.  The “floating patch” behavior, where there is systematic movement of  patches across the landscape, is therefore likely bring in new species (included many exotics to the region) forming a synthetic community, which is constantly adapting to the changing balance of interior and edge habitat. The species that are considered to be endangered through loss of old-growth forests will likely disappear at a much faster rate with the IU system of management. Further, new species may significantly alter the fuel structure within the forest (McDonald et al. 1989; Panetta and Hopkins 1991) leading to increased fire hazard.  141 Dispersed harvesting is also associated with changes in microclimatic conditions such as wind. Wind can be responsible for direct damage in the form of either wind pruning (Caborn 1957) or windthrow of trees (Saunders et a!. 1991). Trees near the edge of recently isolated patches are more prone to the windthrow risk, as they have matured within a closed canopy and have therefore grown in the absence of wind and lack the necessary support mechanism to deal with it. The creation of gaps created by windtbrow is likely to further extend the edge influence into interior habitats. For some time there has been debate as to whether the most appropriate strategy for biodiversity conservation should be to protect a Single Large reserve Or Several Small reserves (the so-called SLOSS debate) (Simberloff and Abele 1984; Gilpin and Diamond 1980; Higgs and Usher 1980; Simberloff 1986).  The IU system will not  provide either of these, as it can only have small patches that support a synthetic biotic community with high mobility.  It will be difficult to control both the external and  internal influences on the patch. On the other hand, the SU system will help to create a network (or a satellite) of fairly large patches of remnants that can be linked to (or around) the protected area network. In the SU system it should be possible to manage external influences by complementing strategies in the timber zone and by having special strategies to manage the internal dynamics of reserve areas and to maintain the patches as natural as possible. Thus, instead of debating SLOSS, it should be possible to have a system of Single Large reserves (representing the protected areas) 4nd Several Small reserves from the multiple use forest areas (SLASS). The large reserves in the SLASS  142 would consist of the large reserves in the protected area network, while the small reserves would be scattered throughout the working forest. In many countries, most of the vast expanses of forests have already been reduced to patches of remnants, and they have the burden of identifying patches of practical conservation value and then managing them to retain their value. Fortunately, British Columbia still has vast tracts of natural forests where representative patches could still be selected outside the protected area network to ensure connectivity of the forest ecosystems. The SU system will help this management strategy.  8.4 IMPLICATIONS FOR VISUAL QUALITY The analysis shows that the opportunity cost of maintaining visual quality is very high. If the SU system of management is adopted it may be possible to select timber zones away from areas that have a very high visual quality requirement. This will meet the twin objectives of meeting the timber supply requirement as well as maintaining visual quality in critical areas. This choice is not available with the IU system as many of the resource emphasis rules lead to a shortage of timber areas. Furtfier, the fragmented landscape caused by dispersed felling under the IU system is likely to negatively impact the beauty of the landscape. Recreation is an important use of the forests. This research did not focus on this topic due to a lack of data in the area concerned. Like visual quality, some forms of recreation such as hiking and cross-country skiing also are in conflict with timber production. The analysis of these uses is likely to produce results similar to that of visual quality.  143 8.5 IMPLICATIONS FOR FOREST STEWARDSHIP In British Columbia, 96% of the forest lands are publicly owned (Forestry Canada 1993). They are managed through a system of crown forest tenures. These tenures are a means of ensuring private management of public property to achieve efficient allocation of resources in the context of the market system. The tenures give property rights for specific uses of forests to the tenure holders. The term “property rights” refers to the entire range of rules, regulations, customs and laws that define rights over appropriation, use and transfer of goods and services (Kula 1992). In other words, they are socially sanctioned and enforceable claims of individuals or groups to the benefits (pecuniary or non-pecuniary) flowing from the property (Haley and Luckert 1995). At the time the current tenure system was designed in British Columbia the forests were seen only as trees and the primary objective of forest management was to liquidate the old-growth as a means of generating direct revenue and providing stable regional employment (Pearse 1988). Hence, the tenure arrangements mostly evolved around the transferring of the right to harvest timber. As stated in my introductory chapter, seeing the forests as only trees has radically changed and now the public views the forests as part of the natural ecosystem that is required for the sustenance of all life on Earth. At the same time, the social values of many of the products produced by the forest have increased tremendously. Successive governments have responded to the call by the public for management of the forest for multiple uses, by modifying public forest policies.  However, the  underlying framework for management of forests through timber tenures has essentially,  144 remained intact. The production of other values has been achieved mainly through a system of command and control type regulations on the harvest and management of timber which have increasingly attenuated existing property rights, and, consequently, have reduced benefits derived by tenure holders. Tenure holders now tend to manage the forests for other values only to the minimum required standards and at the least cost. This no longer encourages the efficient allocation of resources as originally intended. The system has now deteriorated to such an extent that the goals of the public and that of the tenure holders are no longer coincidental. Originally, when the tenures was designed, it was thought that the achievement of private goals would also, to a great extent, help achieve public goals. But now, with changed social values, many of the initiatives taken by tenure holders to achieve their objectives lead to negative social effects (Haley and Luckert 1995). Beside the lack of management for non-timber resources (other than that required by regulation), there is evidence to indicate that tenure holders do not even invest adequately in measures to ensure the continued timber productivity of their leasehold lands (Luckert 1988; Zhang 1994). This is because the tenures do not grant comprehensive property rights to the growing and harvesting of timber. Currently there is a pressing need to re-examine the forest policy of British Columbia and restructure the economic instruments embodied in the forest tenure system (Haley and Luckert 1995). The adoption of the SU system and the equitable allocation of timber zones among tenure holders would help solve many of the problems associated with the current tenure systems. Timber zones as a form of tenure automatically confer some amount of security (for example, no withdrawals, limited harvesting regulations).  145 One of the reasons why governments are reluctant to give full security to tenure holders is because this would involve sacrificing of much of the flexibility considered necessary to deal with inevitable changes in social values and aspirations. The wisdom and foresight of the early forest planners is now evident with the recent radical change in the social system of values. Managing limited areas of forest lands as timber zones (similar to agricultural land) is less likely to jeopardize the choice of future generations. Timber zones would be useful in creating perpetual, fully transferable and secure tenures as there is less likely to be much need for attenuation of rights to protect and provide for the production of other values.  Such tenures for single use would not only simplify the  responsibilities of tenures holders and reduce management costs but are also more likely to provide incentives for intensive timber management. Pearse (1988) examined the possibility of extending rights granted over forest land to include other values, particularly wildlife, fish and some forms of recreation. Under the SU system of management, it should also be possible to grant limited property rights for a combination of marketable non-timber goods and services produced from forests so that the tenure holders could manage for an optimal mix.  These lands should  be selected away from forest land required for the generation of critical non-marketable values. This type of allocation system is not feasible under the current IU system. Under the SU system of management, the forests that have been allocated for the production of non-marketable goods and services could be directly managed by public agencies in the pursuit of social objectives. This system of management is very common in countries like United States of America, France, Germany, New Zealand, and in many  146 developing countries. Such a system would prevent conflicts between private and public interests and thereby avoid high transaction costs involved in protecting the public interests in privately managed public forests (Haley and Luckert 1995).. There has been an increasing level of enthusiasm amongst British Columbians for the establishment of Community forests (Haley and Luckert 1995).  Management of  forests through a system of single and integrated use zones, as proposed in this study, may help in developing simple tenure arrangements for community forests.  Depending  on the interests of the community, Crown tenures could be designed with fairly secure and comprehensive property rights for specific uses relating to the single use zones (timber or some forms of recreation) or to the integrated use zones (managing for nontimber multiple uses). According to Haley and Luckert (1995), one form of tenure will not serve today’s varied and frequently conflicting public objectives. They suggest a system of diverse but complementary tenure arrangements that will depend upon the mix of values the land in question is expected to produce.  This type of tenure reform may not be easy to  implement with the current practice of integrated management. On the other hand, the adoption of a system of single uses and complementary integrated uses as proposed in this research will facilitate such reforms.  8.6 IMPLICATIONS FOR FOREST RENEWAL PLAN The BC Forest Renewal Act (SBC Chap.3 Vol. 1 Bill 32, 1994) has an ambitious plan to rehabilitate neglected forest areas. One of the main goals of this plan is the renewal of the forest with major investments in replanting, spacing, pruning and  147 fertilization. Currently, some timber harvests are taking place in marginally productive lands. These harvests fall within the extensive margin because of the accumulation of many years of growth. Replanting and intensive silviculture in some of these marginally productive areas will not be economically feasible. Rehabilitation of these areas will require a strategy different from that of simply increasing timber productivity. About 80% of the so-called Not satisfactorily Restocked (NSR) lands in BC (874,000 hectares in 1988) is found on poor and medium sites (Ministry of Forests 1988). In 1993, the extent of NSR lands is estimated to be 1,362,407 hectares (Ministry of Forests 1994).  For purposes of allocating funds for restoration of the NSR lands,  Thompson et al. (1992) ranked all NSR sites according to their cost! benefit ratio. Restoration in this study meant the restoration of timber productivity of the land. Costs refer to cost of planting and benefits refer to financial benefits accruing from harvest of commercial timber. Except for spruce on good quality sites near mills, all of the species types when established in poor and medium sites are estimated to have a benefit cost ratio of less than one at a discount rate of 1.5%. In fact, the performance of many species types on good sites also show a benefit cost ratio of less than one. In order to utilize the available funds for restoration, Thompson et al. (1992), recommended that the sites that show a benefit!cost ratio of 0.81 or greater should be considered for reforestation. This is based on the assumption that increasing timber productivity will also enhance other non-timber values and therefore increase the actual benefit! cost ratio of the site. This is questionable as this research shows intensive silviculture practices under the IU system is  148 likely to bring about more damage to the natural environment in terms of fragmentation of interior habitats. Timber value should be enhanced only when it is economically feasible to do so and this should be done only on areas of NSR lands that could be permanently maintained as timber zones. NSR lands other than those identified for timber zones should be rehabilitated by other means designed to enhance their intrinsic values (possibly based on priorities for wildlife and visual quality enhancement). In these areas, the type of species selected for reforestation and the method of reforestation may be quite different from those used for timber production.  Carrying out intensive silvicultural practices to  increase timber productivity under the IU system framework may negatively impact such ecosystems and, consequently, many social values.  8.7 IMPLICATIONS FOR SHORT TERM TIMBER SUPPLY Currently, much of British Columbia’s forest is locked up under various constraints and forest harvesting is extending towards its extensive margin or even beyond. As harvesting moves towards its extensive margin two things happen. First, the cost of harvesting increases and second, more roadless areas will be harvested leading to the disturbance of these habitats. Adoption of the SU system will confer several types of immediate benefits to society.  First, it will release for immediate harvest substantial areas of productive  operable land, hitherto locked up under adjacency and cover constraints. The benefits from harvesting these newly released areas are likely to be higher than that of the marginal lands that are currently harvested. This assertion is based on the assumption  149 that relatively more productive areas were harvested in the past than are being harvested at present and that the relaxation of adjacency constraints would release productive areas carrying high quality timber which had been set aside in the past. The delivered wood cost of timber from these areas is also likely to be lower due to shorter hauling distances and easier accessibility. Second, the SU system, unlike the IU system, will liquidate old-growth faster and, therefore, allow the modified forest with higher volume productivity to be harvested earlier.  Intensive silviculture such as spacing and precommercial thinning, though  postponing the age of culmination of MAI, will ensure early availability of high value merchantable timber. Rotation ages for the modified forests can, therefore, be lowered based on the window of economically feasible ages. By carefully planning these harvests over the faildown period, the falldown effect can be averted or mitigated. Third, the effect of falldown can be mitigated to some extent by timing commercial thinnings during the fall down period in the SU system. Though this could be done in the IU system too, the accompanying negative environmental impacts may not justify such practices. Fourth, the SU system will lead to immediate savings in expenditure. This will come from three sources: i) savings in delivered wood cost; ii) savings in silviculture expenditure; and iii) savings in administrative costs. Fifth, with the adoption of the SU system, the extension of the margin of operable areas into roadless areas can be halted. This will release more areas for non-timber uses and substantially reduce resource use conflicts.  150 8.8 STRATEGY FOR THE FUTURE  This research has analyzed the impacts of current resource emphasis rules in British Columbia on timber supply and on the environment.  The results and the  foregoing discussion indicate that the SU system can generate higher rents with less environmental impact.  Therefore, one way of ensuring continuous timber supply at  reasonable prices, while maintaining ecosystem integrity, is the adoption of the SU system. The bulk of timber production in the future should be from tree farms dedicated solely for that purpose.  This will encourage the practice of intensive silviculture  including genetic improvements and the application of other types of bio-technology. This is very much like intensive farming of agricultural lands. Areas that require special protection within and outside the timber zones, such as riparian areas and environmentally sensitive areas, should be zoned and special resource emphasis rules should be drawn up to protect them. Other than timber production and a few recreation activities such as snowmobiling, production of most of the other multiple uses of forests do not conflict with each other. In these cases, timber could be harvested when it is seen to be complementary to the production of other goods and services. Having such a mosaic of single use zones and multiple use zones in a management unit may appear to be a SU system but at a regional level it is really a form of integrated management of multiple use forests. Crown forest tenures through which the private sector participates in the management of the public forests could be restructured as suggested by Haley and Luckert (1995) to meet the provincial objectives of sustaining the  151 natural ecosystems and the economy they support. In this way the efficiency of forest management can be improved.  152  9 SUMMARY AND CONCLUSIONS This study deals with multiple use management in forestry. The main objectives of the study were: i) to review the literature on the economic theory of multiple use and examine various approaches taken by foresters to the practice of multiple use forestry; ii) to estimate the impact, in terms of timber supply and rent, of integrated management practices designed to maintain a specific quality of wildlife habitat and visual aesthetics; iii) to develop alternative land use systems that incorporate zones ‘specializing in the production of timber (timber zones); iv) to compare with respect to timber supply, rent and selected environmental indicators, the integrated use management system with the alternative systems emphasizing timber zones; v) to estimate the impact (with respect to timber supply, rent forgone and selected environmental indicators) of intensive timber management on both systems; and vi) to estimate the impact of enhancing wildlife habitat and visual aesthetics on timber zones in terms of timber supply and rent forgone. The literature review shows that the benefits of multiple use forests can be measured in terms of timber rent and a few environmental indicators (or landscape statistics). When the price of timber is constant, rent is dependent on the level of harvest, the delivered wood cost and the administrative costs of forest management. Since there is an inverse relationship between the flow of non-timber goods and services (amenity services), the higher the flow of amenity services demanded, the lower the timber rent, and vice versa. Current forest management practices in British Columbia, except for zoning for parks and protected areas, emphasize integrated management where each  153 hectare of forest land is expected to be managed for multiple use. Under this system of management the flow of amenity services through time is fixed by enforcing resource emphasis rules which determine the harvestable volume of timber. When delivered wood costs, administration costs and the price of timber are constant the harvest volume will determine the value of timber rent.  The literature review also shows that certain  peculiarities (such as differences in site productivity, diseconomies of jointness in production and the pattern of response to management efforts) exhibited by multiple use forests may favor specialization in the production of timber. Economic theory suggests that when management is constrained by the production of a fixed flow of amenity services, multiple use systems that emphasize special timber productibn zones generate a higher timber rent than systems that emphasize integrated production of all uses at the same time from each hectare of land. These theoretical findings were empirically tested by simulation modeling with ATLAS and SIMFOR. For the empirical study, the Akolkolex drainage, which is a sub-unit of the Reveistoke Timber Supply area in the Reveistoke Forest District of British Columbia was selected. Reveistoke has severe resource use conflicts, and the management strategies developed by the District to deal with this problem are fairly well advanced when compared to other districts in British Columbia.  This District has designed special  harvesting patterns and rules to maintain a continuous flow of visual quality and wildlife habitat values throughout the planning horizon. This study shows that the opportunity cost of current management practices on a 120 year planning horizon is about 60% of the potential timber supply and about 46  154 million dollars (at a 3% discount rate) in foregone rent, which is equivalent to a loss of 123 dollars per hectare per year in present terms. The opportunity costs of wildlife and visual quality management per year for the whole timber supply area amount to 8.8 million dollars and 6.2 million dollars, respectively. In other words, these are the trade offs involved in managing for non-timber values. A major cause of the reduction in timber supply and rent in these practices is the adjacency requirement. Two questions should be raised in this context: i) Can these opportunity costs be justified in terms of resulting benefits? and ii) are there other ways of managing land to produce the same or an enhanced mix at lower cost?. Two types of multiple use systems, referred to as the Integrated Use (IU) system and the Single Use (SU) system were devised. The IU system treats the whole operable area of Reveistoke 1 as an integral production unit where timber is produced as one of the multiple uses, with all spatial and temporal constraints in place. In the SU system, a portion of the operable area (46% of Reveistoke 1) is allocated as a single use area, specifically for timber production, while the balance is allocated for non-timber multiple uses. The study indicates that with unconstrained timber production, it should be possible to produce the same evenflow volume currently possible through integrated management within Reveistoke 1 from 46% of the net productive area of the unit.  This  will effectively release 54% of the forest lands solely for the production of values other than timber.  155 Two planning horizons: i) 120 years, and ii) 240 years were investigated. Both of these systems at basic levels of management (that is, relying only on natural regeneration for restocking) produce almost equivalent evenflow volumes of timber at each of the planning horizons.  In both cases, it was found that going from a 120 year to a 240 year  planning horizon reduced the sustained timber supply (evenflow volume). For example, in the IU system there was a reduction in evenflow volume from 14,000 m /year to 3 11,800 m /year. 3 The impact of intensive timber management at three levels (basic, medium and high) on timber supply, rent and a few selected environmental indicators (density of roads, distribution of seral stages in the planning horizon, distribution of ecosystems within seral stages, percent edge habitat, percent of regeneration area affected by edge, and distribution of patch sizes in very old-growth) were investigated for both land use systems on a 240 year planning horizon. Intensive timber management increases both the quantity and quality of timber harvested. At medium and high management intensities, evenflow volume from the SU system is increased by approximately 38%. For the IU system, the evenflow value rises by approximately 32% under medium intensity management and by 25% at high intensity. The drop at high intensity is the result of additional adjacency requirements caused by thinnings. Rent  (@ 2% discount rate) from the SU system under medium and high intensity  management is approximately 216% relative to the basic IU system. This reflects the increased quality of timber harvested. This relative value is as high as 444% when 0% discount rate is used. The beneficial effects of intensive management are not as obvious  156 at high discount rates, because the benefits occur during the second half of the planning horizon and are heavily discounted. While the choice of discount rate drastically affects the selection of management intensity, it does not alter the superior performance of SU over IU systems (Figures 30 and 31). Sensitivity analysis of timber rent to 12% changes in price show that the SU system is less sensitive than the IU system, and their relative performance does not change. The timber zone selected in this study consists of nearly 11.5% poor sites by area. If the timber zones were selected from good and medium sites only, then the evenflow volumes and the rent would be proportionately higher. By practicing intensive timber management it is also possible to further reduce the minimum area of the timber zone required to produce an evenflow of volume of timber equivalent to what the IU system would provide at basic intensity. With medium and high intensities (on a 240 year planning horizon), the timber zone could be reduced from 46% to 35% of the net Revelstoke 1 land base. The minimum area of the unconstrained timber zone required to produce rents equivalent to those produced by the IU system at basic intensity could range from 10%  (@ 0% discount rate) to 50% (@4% discount rate)  of the net area of Revelstoke 1. At high discount rates, long term benefits are heavily discounted and, therefore, more area is required. Currently the road system cost is only about 5% of the delivered wood cost. But this is likely to increase severalfold if road deactivation and reactivation costs are included. These costs are likely to be less with the SU system, as a permanent road network will be maintained for intensive timber management operations. Further, higher  157 quality roads will be built within the timber zone thus reducing detrimental environmental effects. Hauling costs may be lower under the SU system because timber has to be hauled for shorter distances. More research is required in this area. Environmental indicators suggest that, in addition to higher volumes and rent produced by the SU system compared to the IU system at varying intensities of management, the SU system offers higher environmental quality. Road density in the SU system is only 65% of that of the TU system. Roads are considered to be one of the key factors contributing to the destruction of pristine environments by increasing slope failure, erosion, siltation, and accessibility for hunters. If the IU system is practiced, then all the forests other than the Protected Areas will be dotted with deactivated and reactivated roads and ecosystem functions will be compromised everywhere. The SU system appears to be one of the best ways of protecting much of the environment, thus guaranteeing high option and bequest values. Modeling of landscape pattern response to harvesting practices under the IU and the SU system shows distinct differences in some of the key landscape indices.  The  study shows that under both the SU and the IU systems, the extent of very old-growth stands (>240 years) as a percent of the landscape during the 240 year planning horizon increases from 28% at the start of the period to 70% at the end. In both cases, nearly 50% to 63% of the original old-growth (>120 years) found at the start of the planning period remains unharvested at the planning horizon. Even within the timber zone, nearly 35% of original old-growth stands remain unharvested.  This is due to the evenflow volume  restrictions on harvesting. It would be wiser to manage the SU system on an area basis  158 rather than by volume.  This would reduce many uncertainties surrounding evenflow  management, especially short term timber supply shortages. The results show that the degree of fragmentation of the interior habitats constituted by very old-growth stands (>240 years) and the percent of edge habitats in stands greater than 120 years are very much higher in the IU system than in the SU system. The degree of fragmentation and the percent of edge habitats increases with the intensity of management and are relatively high for the IU system.  Presently, a major  concern of conservation biologists is the accelerating disappearance of interior natural habitats, particularly those associated with natural old-growth areas. The relatively high degree of fragmentation of interior habitats and the high percent of edge habitats in the IU system is an indication of the loss of valuable interior habitat.  Thus, by practicing the  current form of integrated management we may be further endangering critical wildlife habitats. The regeneration of pioneer, light-demanding species may be seriously affected by microclimatic conditions created by old-growth stands surrounding the regeneration areas.  For the IU system at various intensities of management, the affected area ranges  from 42% to 45% of the regeneration area. While for the SU system it is as low as 10% to 14%. Implications to stand growth and yield are unknown and were not considered in this study. Maintaining visual quality has high opportunity costs in terms of timber rent foregone yet has no direct bearing on ecosystem function. The decision to maintain this quality at specific levels should take into account the social costs involved. In the SU  159 system, it is possible to avoid visually sensitive areas when there is a possibility of locating timber zones elsewhere. Thus, it may be possible to generate high timber rents while adequately protecting visual quality. This may not be possible with the IU system as there is a requirement that every hectare in the system be managed for multiple use. Enhancement of non-timber values within the timber zone reduces the evenflow volume of timber. As for rent, except for visual quality with partial retention, equivalent or more rent could be generated at medium and high management intensity levels even with the enhancement of selected non-timber values. My research findings have several short term implications. The adoption of the SU system would release many highly productive areas with high value timber hitherto locked up in adjacency and cover constraints for short term harvests. The delivered wood costs and road costs from these released areas will also be very much lower than that of current areas due to shorter hauling distances, easy accessibility, and the advantage of using existing roads. Intensive silviculture such as spacing and precommercial thinning, though postponing the age of culmination of the MAT, may make it economically feasible to harvest the available high value timber at an earlier age (shorter rotation). Though the same thing is feasible with the TU system, such strategies are accompanied by various environmental problems.  Adoption of such intensive management in the SU system  would be likely to mitigate the faildown effect as these harvests can be timed to the periods of falldown. There will be substantial savings from avoiding intensive silviculture such as planting and spacing in areas other than designated zones. More research is required to fully evaluate this aspect.  160 This research helps to estimate the trade-offs involved in producing non-timber values in alternative land use systems.  Policies which attempt to practice intensive  timber management within an IU system are likely to irreversibly modify all forests outside protected areas and parks throughout the province. The SU system, on the other hand, offers protection to forested areas outside the protected area system and provides for more future choices. Areas that require special protection within and outside the timber zones, such as riparian areas and environmentally sensitive areas, should also be zoned and special resource emphasis rules should be drawn up to protect them. Timber production, in specialized zones for other uses and in multiple use zones, could be practiced to a level that is found to be complementary. Having such a mosaic of single use zones and multiple use zones in a management unit may appear as a SU system at the scale of a management unit, but at a regional scale they are really a form of integrated management with a higher degree of efficiency. Burton (1995) suggests a land use management system that incorporate elements of both the SU system and the JU system. He suggests an allocation ratio of 1: 2: 1 for protected areas, integrated use and timber zones. This type of allocation appears to be a good solution to the highly polarized land use conflicts of the present and provides choices for the future generations. Timber zones need not be selected as a single contiguous unit. Several timber zones could constitute a single sustained yield unit. Based on land prductivity and economic feasibility, several areas can be defined as timber zones. It would be ideal if timber zones were selected in every biogeoclimatic sub zone, thus utilizing the  161 comparative advantage afforded by natural productivity of these sites.  Species  composition of modified forests would vary with each timber zone based on its biogeoclimatic subzone. This will also give a mixture of timber products for a sustained yield unit.  Identification of timber zones should be based on ecology as well as  economics. First, there should be selection of productive sites in different biogeoclimatic subzones for different species that will constitute the modified forest. As far as possible, timber zones should only be on good and medium sites. Care must be taken to ensure that road density and fragmentation of the forest is not further increased. This should be followed by identification of economically feasible sizes of timber zones. Rouck and Nelson (1994) have shown that the size of the Reveistoke Timber Supply area (TSA) could be partitioned into four sustained yield units without significant adverse impacts on volume flow.  This type of area-based research should be extended to identifS’  economically feasible timber zones that will constitute a sustained yield unit. All timber production zones need not be barren with respect to wildlife and visual quality.  Depending on their location and importance to wildlife habitats and visual  quality, these resource emphases could be enhanced to varying degrees. This will be an important consideration if the timber zone is selected close to a park or a critical wildlife habitat. The trade-offs involved in each case have been established in this study. For instance, it will be possible to accommodate or enhance certain categories of wildlife in timber zones without having the adjacency constraint. A blanket requirement to have the  162 adjacency constraint in all timber producing areas, as is currently practiced, has a very high opportunity cost and there is little justification to do so. Further research will have to be directed towards this topic. An important economic benefit of establishing timber zones may be a reduction in management and transaction costs. Integrated management requires major investments in knowledge and presents complex management problems which increase administration costs substantially. Furthermore, in the case of crown lands which are managed by the private sector, monitoring costs can be very high under integrated management. Zoning for timber production can substantially reduce the expenditures, thus allowing more funds (essentially from savings) to be diverted to the management of non-timber producing zones. In British Columbia, all forest tenures provide holders with exclusive right to harvest timber but no opportunities to benefit from the production of other forest products (Haley and Luckert 1990). Externalities arising from multiple use of forest land are dealt by means of command and control type of regulations (Haley and Luckert 1995). This gives the tenure holder incentives to manage the forest for other values to the required minimum standards at least cost, thus jeopardizing the optimal allocation of resources. The adoption of the SU system with equitable distribution of timber zones among tenure holders may reduce the problems caused by regulations attenuating the tenure rights. Timber zones will be useful in creating perpetual fully transferable and secure tenures. Complementary tenure arrangements (Haley and Luckert 1995) based on the mix of  163 values the land in question is expected to produce can be developed for the other zones in the SU system. This is likely to help in the optimum allocation of resources in a market based economy. The results of this study also have implications for the Forest Renewal Plan. It is inefficient to invest forest renewal funds on areas destined for integrated management because of the high opportunity cost involved in accommodating non-timber resources. These funds could be better utilized in single use timber zones. The same is true for restocking the not satisfactorily restocked lands (NSR) in the province. Only NSR lands that come within the timber zone should be restocked. Restocking of other NSR lands should be based on priorities for wildlife and visual quality enhancement. In these cases the type of species selected for reforestation and the method of reforestation may be quite different from those used for timber production. That is, reforestation should be based more on the principles of restoration ecology than on those of silvicultbre. This study is a step towards achieving the dual provincial goals of sustaining ecosystems and the natural resource based economy. It establishes a decision support framework needed to analyze some aspects of this complex problem. This framework is transportable to other areas. While certain resource issues (e.g., water quality) may be more significant elsewhere, the decision support framework developed here can be readily adapted to meet specific issues. Most importantly, this study demonstrates the strong potential of single use management through land use zoning as a strategy to balance economic and  164 environmental values from our forests.  It offers an innovative method to achieve  sustainability of timber and non-timber resources.  165  BIBLIOGRAPHY Alverson, W.S., W. Kuhlmann and D.M. Wailer. 1994. Wild Forests. Conservation Biology and Public Policy. Island Press. Washington, D.C. Bowes, M.D and J.V. Krutilla. 1985. Multiple use management of public forest lands. In: A.V. Kneese and J.L Sweeney (Editors), Handbook of Natural Resource and Energy Economics, 11:531-543. Elsvier Science Publishers, Amsterdam. Bowes, M.D. and J.V. Krutilla. 1989. Multiple-Use Management: The Economics of Public Forestlands. Resources for the Future. Washington, D.C. 357 p. Bradshaw, F.J. 1992. Quantifying edge effect and patch size for multiple-use silviculture a discussion paper. Forest Ecology and Management 48: 249- 264.  -  Braumandl, T.F. and M.P. Curran. 1992. A Field Guide for Site IdentfIcation and Interpretationfor the Nelson Forest Region. B.C. Ministry of Forests. Victoria. Brundtland, G.M. 1987. Our Common Future. The World Commission on Environment and Development. Oxford University Press. 4.00 p. Burton, P.J. 1995. The mendelian compromise: A vision for equitable land use allocation. Land Use Policy 12(1):63-68. Caborn, J.M. 1957. Shelterbelt and Microclimate. Forestry Commission Bulletin 29. Her Majesty’s Stationary Office, Edinburgh, Scotland. 135 p. Cale, W.G., G.M. Henebry, and J.A. Yeakley. 1989. Inferring process from pattern in natural communities. BioScience 39: 600-605. Calish, S., R.D. Fight, and D.E. Teeguarden. 1978. How do non-timber values affect Douglas-Fir rotations? Journal ofForestry 76: 217-221. Chen, J., J.F.Franklin and T.A. Spies. 1992. Vegetation edge response to edge environments in old-growth Douglas fir forests. Ecological Applications 2: 387-396. Clements, S.E., Dallain, P.L. and M.S. Janmick. 1990. An operational, spatially constrained harvest scheduling model. Canadian Journal ofForest Research 20: 143 81447. Cline, W.R. 1992. The Economics of Global Warming. economics. Washington, D.C. 399 p.  Institute for international  166  Cropper, M.L. and P.R. Portney. 1991. Discounting and the evaluation of life saving programs. Journal ofRisk and Uncertainty 3: 369- 379. Dahlin, B. and 0. Salinas. 1993. Harvest scheduling under adjacency constraints- a case study from the Swedish sub-alpine region. Scandinavian Journal ofForest Research 8: 281-290. Dana, S.T. 1943. Multiple use, biology and economics (Editorial). Journal ofForestry 41:625-626. Dasmann, R.F. 1964. Wildflfe Biology. J.Wiley & Sons. New York. 231 p. Daust, D.K. 1994. Biodiversity and Land Management: From Concept to Practice. MSc Thesis, University of British Columbia, Vancouver. 99 p. Daust, D., C. Galindo-Leal and F. Bunnell. 1993. Predicting Impacts ofForest Operations on Biological Diversity. Unpublished report. Centre for Applied Conservation Biology, Faculty of Forestry, UBC. 29 p. Davis, L. S. and K. N. Johnson. Company, New York. 790 p.  1987.  Forest Management.  McGraw- Hill Book  Faustman, M. 1849. On the determination of the value which forest land and immature stands possess for forestry. In: M. Gane ed., Martin Faustman and the Evolution of Discounted Cash Flow. Institute paper No. 42, Commonwealth Forestry Institute (University of Oxford). 1968. Feldstein, M. S. 1964. The social time preference rate in cost-benefit analysis. Economic Journal 74: 360-379. Forestry Canada. 1993. Selected Forestry Statistics Canada. Economics Directorate. Information E-X-46. Ottawa, Ontario.  1992.  Policy and  Forman, R.T.T and M. Godron. 1986. Landscape Ecology. John Wiley & Sons. New York. 6l9p. Giipin, M.E., and J.M. Diamond. 1980. Subdivision of nature reserves and the maintenance of species diversity. Nature 285: 567-568. Government of B.C. 1993. A Protected Areas Strategy for British Columbia. protected areas component of B.C.’s land use strategy. Victoria, B.C. .38 p.  The  Gregory, G.R. 1955. An Economic Approach to multiple uses. Forest Science 1:5-18.  167  Gregory, G.R. 1972. Forest Resource Economics. New York: Ronald Press Company. 548 p. Gregory, G.R. 1987. Resource Economics for Foresters. John Wiley & Sons, New York. p. 477 Guttierrez,R.J., Carey, A.B. eds. 1985. Ecology and Management of the Spotted Owl in the Pacific NorthWest. General Technical Report. PNW_185. USDA Forest Service. Pacific NW Forest Range Exp. Station. Portland, Oregon. Hagenstein, P.R. and B. Dawdle. 1962. A theoretical method for examining forest land use alternatives. Journal ofForestry 60: 87-91. Haley, D and M.K Luckert. 1995. Tenures as Economic Instruments for Achieving Objectives of Public Forest Policy in British Columbia. Prepared for an executive workshop on economic instruments for the protection of forest reources. Faculty of Law, University of Victoria. June 22. 36 p. Hampson, S.F. 1972. Highland forestry: an evaluation. Journal ofAgricultural Economics 23: 49-57. Hansen, A.J., S.L. Garman, P. Lee, and E. Horvath. 1993. Do edge effects influence tree growth rates in Douglas-fir plantations? Northwest Science 67 (2): 112-116. Hartman, R. 1976. The harvesting decision when a standing forest has value. Economic Inquiry 14:52-58. Hartman, R.W. 1990. One thousand points of light seeking a number: A case study of CBO’s search for a discount rate policy. Journal of Environmental Economics and Management 18: S3-S7. Heaps, T. and B. Pratt. 1989. The Social Discount rate for Silvicultural Investments. FRDA Report 071. Forestry Canada, Victoria. Higgs, A.J., and M.B. Usher. 1980. Should nature reserves be large or small? Nature 285: 568. Jacobs, M. 1993. The Green Economy: Environment, Sustainable Development and the Politics ofthe Future. UBC Press, Vancouver. 312 p. Johnson, K.N. and S. Crim. 1981. Integrating Long Term and Short Term Scheduling in Forest Planning. Unpublished paper. 20 p  168  FORPLAN Version 2: Mathematical Johnson, K.N. and T.W. Stewart. 1987. Programmer’s Guide. Land Management Planning Systems Section Report, USDA. Forest Service, Washington, D.C. Kimmins, J.P. 1994. Sustainable development in Canadian forestry in the face of changing paradigms. Pages 3 -18. In: Proceedings of conference on “Timber Supply in Canada: Challenges and Choices”. Natural Resources Canada. Ottawa. Kula, E. 1992. Economics ofNatural Resources and the Environment. Chapman & Hall, London. 284.p. Laing & McCulloch. 1993. Background Report for the Development of Juvenile Spacing Guidelines in British Columbia. Report prepared for the B.C. Ministry of Forests, Silviculture Branch, Victoria, B.C. Leuschner, W. A. 1990. Forest Regulation, Harvest Scheduling, and Planning Techniques. John Wiley & Sons Inc., New York. 281 p. Liu, G. 1994. Evolution Programs, Simulated Annealing and Hill Climbing Applied to Harvest Scheduling Problems. Unpublished M.F. Thesis, University of British Columbia. 59 p. Lockwood, C. and T. Moore. 1993. Harvest Scheduling with Spatial Constraints: A Simulated Annealing Approach. Canadian Journal ofForest Research 23: 468-478. Lovejoy, T.E., R.O. Bierregaard, Jr., A.B. Rylands, J.R. Malcolm, C.E. Quintela, L.H. Harper, K.S. Brown, Jr., A.H. Powell, G.V.N. Powell, H.O.R. Schubart, and M.B. Hays. 1986. Edge and other effects of isolation on Amazon forest fragments, pp 257 285. In: M.E. Soule, editor. Conservation Biology: The Science of Scarcity and Diversity. Sinauer, Sunderland, Masschussetts, USA. -  Lowenstein, G. 1987. Anticipation and valuation of delayed consumption. The Economic Journal 97: 666 -84. Luckert, M.K. 1988. The Effect of Some British Columbia Forest Tenures on the Distribution of Economic Rents, the Allocation of Resources, qnd Investments in Silviculture. Unpublished Ph.D thesis. University of British Columbia. 238 p. Luckert, M.K. and D. Haley. 1993. Candian forest tenures and the silvicultural investment behaviour of rational firms. Canadian Journal ofForest Research 23: 1060- 1064. 1989. Wildlife Macdonald, L.A.W., L.L. Loope, M.B. Usher, and O.Hamann. conservation and the invasion of nature reserves by introduced species: a global perspective. Pages 215- 255. In: J.A.Drake, H.A. Mooney, F.di Castri, et al. editors. Biological invasions. A global perspective. John Wiley & Sons. New York.  169  Marglin, S. 1963. The opportunity cost of public investment. Quarterly Journal of Economics 77: 274-289. Meidinger, D. and J.Pojar. 1991. Ecosystems of British Columbia. B.C. Ministry of Forests. Victoria, B.C. 330 p. Ministry of Forests. 1988. Unpublished report quoted in paper by Thompson, W.A., P.H. Pearse, G.C. Van Kooten and I. Vertinsky. 1992. Rehabilitating the backlog of unstocked forest lands in British Columbia: A preliminary simulation analysis of alternative strategies, pp 99-130. In: P.N. Nemetz, editor. Emerging Issues in Forest Policy. UBC Press, Vancouver, B.C. Ministry of Forests. 1993a. Revelstoke TSA timber supply analysis. Resources Branch, Ministry of Forests, B.C. 59 p.  Integrated  Ministry of Forests. 1993b. VDYP interactive application. User Guide. Version 4.5. Inventory Branch, Ministry of Forests, Victoria. 31 p. Ministry of Forests. 1994. Forest, Range & Recreation Resource Analysis. Ministry of Forests, Victoria. 308 p. Mishan, E.J. 1976. Cost-Benefit analysis. New and Expanded Edition. Publishers. New York, U.S.A. 454 p. Mitchell, K.J. 1975. Monograph 17.  Dynamics and Simulated Yield of Douglas-fir.  Praeger  Forest Science  Mitchell, K.J., R.M. Kellogg, and K.R. Poisson. 1989. Siiviculturai treatments and end product value. In: R.M. Kellogg. editor. Second Growth Douglas fir: Its Management and Conversion for Value. A Report of the Douglas fir Task Force, Forintek Canada Corp., B.C. Munasinghe, M. 1993. Environmental Economics and Sustainable Development. World Bank Environmental Paper Number 3, The World Bank. Washington, D.C. 112 p. Murray, A.T. and R.L. Church. 1993. Heuristic solution approaches to operational forest planning problems. OR Spectrum. 20 p. Nelson, J.D. 1994. Economics of Partial Cutting Relative to Clear Cutting. Unpublished paper. Forest Operations, University of British Columbia. 5 p. Nelson, J.D. and Brodie, J.D. 1990. Comparison of a random search algorithm and mixed integer program for solving area based forest plans. Canadian Journal of Forest Research 20: 93 4-942  170  Nelson, J.D. and S.T. Finn. 1991. The influence of cut-block size and adjacency rules on harvest levels and road networks. Canadian Journal ofForest Research 21(5): 595-600. Nelson, J.D., and A.F. Howard. 1991. A three-stage heuristic procedure for allocating spatially and temporally feasible cutting rights on public lands. Canadian Journal of Forest Research 21:762-768. Nelson, J., and Liu,G. 1994. Simulated annealing and forest planning problems. In: Proceedings of International Seminar on Forest Operations under Mountainous conditions. Pages 28-36. July 24-27. Harbin, P.R. of China. Nelson, J.D., Brodie, J.D and J. Sessions. 1991. Integrating short term logging plans with long term harvest schedules. Forest Science 37: 101-102 Nelson, J.D., M. Hafer and T. Shannon. 1993. ATLAS: A Tactical Landscape Analysis System V 1.3. User Manual. Forest Operations Group, Department f Forest Resources Management, University of British Columbia, BC. 35 p. Nelson, J.D., K. Rouck, and T. Shannon. 1994. Area Based Timber Supply Analysis ofthe Revelstoke Timber Supply Area. Report to the Ministry of Forests, Revelstoke Forest District. Department of Forest Resources Management, University of British Columbia, Vancouver. Newman, D.H. 1988. The Optimal Rotation: A Discussion and Annotated Bibliography. USDA Forest Service, General Technical Report SE-48. Ashville, NC. 4.’7p. Noss, R.F. 1991. Effects of edge and internal patchiness on avian habitat use in an oldgrowth Florida hammock. NaturalAreas Journal 11(1): 34-47. O’Hara, A.J., Faaland, B.H., and Bare,B.B. 1989. Spatially constrained timber harvest scheduling. Canadian Journal ofForest Research 19: 715-724. O’Keef, T. 1990. Holistic (New) Forestry: Significant difference or just another gimmick? Journal ofForestry April:23- 24. Owen, S. 1994. West Kootenay- Boundary Land Use Plan. Commission on Resources and Environment. Victoria, B.C. 162 p, Append. Paine, R.T., and S.A. Levin. 1981. Intertidal landscapes: disturbance and the dynamics of pattern. Ecological Monographs 51: 145-178. Panetta, D., and A.J.M. Hopkins. 1991. Weeds in corridors :invasion and management. Pages 341-35 1. In: D.A. Saunders, and R.J. Hobbs. editors. Nature Conservation: The Role ofCorridors. Surrey Beatty and Sons. chipping Norton, Australia.  171  Pearse, P.H. 1969. Towards a theory of multiple use: a ease of recreation versus agriculture. Natural Resources Journal 12: 56-75. Pearse, P.H. 1988. Property rights and the development of natural resource policies in Canada. Canadian Journal ofPublic Policy. March: 307 320. -  Pearse, P.H. 1990. An Introduction to Forestry Economics. Vancouver: UBC Press. 226 p. Pearson, G.A. 1944. Multiple use in forestry. Journal ofForestiy 42: 243-249. Peel, S. 1991. The Future of Our Forests. Forest Resources Commission, Victoria, B.C. p. 97 Pezzey, J. 1992. Sustainable Development Concepts: An Economic Analysis. World Bank Environment Paper No.2. World Bank. Washington, D.C. 71 p. Poisson, K.N. (Stand Modeling analyst, Ministry of Forests, Victoria) 1994. Personal communication. Price, C. 1993. Time, Discounting and Value. Blackwell Publishers. Oxford, U.K. 393 p. Price, L. and J. Blake. 1993. GIS based multiresource analysis in Reveistoke Timber Supply Area. Pages 227-239. In: GIS 93 Symposium. Vancouver, B.C. Rouck, K.B. and J.D. Nelson. 1994. Timber Supply and Economic Impacts associated with sustained yield unit size. Forestry Chronicle (in press) Sahajananthan. 5. 1994. Single and Multiple Use of Forest Lands in British Columbia: The Case ofthe Reveistoke Forest District. Preliminary Report submitted to the Ministry of Forests, Revelstoke Forest District. B.C. 25 p. Saunders, D.A., G.W. Arnold, A.A. Burbidge, and A.J.M. Hopkins. 1987. The role of remnants of native vegetation in nature conservation: future directions. Pages 387- 392. In: D.A. Saunders, D.W. Arnold, A.A. Burbidge, and A.J.M. Hopkins. Editors. Nature Conservation: The Role of Remnants of Native Vegetation. Surrey Beatty and Sons. Chipping Norton, Australia. Saunders, D.A., R.J. Hobbs, and C.R. Margules. 1991. Biological consequences of ecosystem fragmentation: a review. Conservation Biology 5:18-32. Sedjo, R. A. and M. D. Bowes. 1990. Managing the Forest for Timber and Ecological Outputs on the Olympic Peninsula. Discussion paper ENR92-02. Energy and Natural Resources Division, Resources for the Future, Washington, D.C.  172  1987. Sharpe, D.M., Guntenspergen,G.R., Dunn,C.P., Lettner, L.A., Steams, F. Vegetation dynamics in a southern Wisconsin agricultural landscape. In: Turner, M.G. Editor. 1987. Landscape Heterogeneity and Disturbance. New York: SpringerVerlag. Simberloff, D. 1986. Design of nature reserves. Pages 315- 37. In: M.B.Usher. Editor. Wildlfe Conservation Evaluation. Chapman and Hall. London, England. Simberloff, D. and L.G. Abele. 1984. Conservation and obfuscations: subdivision of reserves. Oikos 42: 399-401. Smith, D.M. 1986. The Practice of Silviculture. Eighth Edition. John Wiley & Sons New York. S2’7p. Stone, M. 1993. An Economic Evaluation of Commercial Thinning Douglas-fir in the Coastal Region of British Columbia. FRDA II Working Paper WP-6-002. MOF, British Columbia. 146 p. Swallow, S.K., P.J.Parks and D.N. Wear. 1990. Policy relevant non convexities in the production of multiple use benefits. Journal of Environmental Economics and Management 19: 264-80. Swallow, S.K., and D.N. Wear. 1993. Spatial interactions in multiple use forestry and substitution and wealth for the single stand. Journal of Environmental Economics and Management 25: 103-120. Tanke, W.C. 1985. PASS- a tool for analyzing alternative harvest schedules. pp. 304-3 14. In: Proceedings of the Society of American Foresters symposium on “System analysis in forest resources”. University of Georgia, Athens. Thomas, J.W., C. Maser, and J.E. Rodiek. 1979. Edges. Pages 48-49. In: J.W. Thomas. Editor. Wildlfe Habitats in Managed Forest: The Blue Mountains of Oregon and Washington. USDA Forest Service Agricultural Handbook Number 553. Thompson, E.F., B. Hallerman, T.J. Lyon, and R.L. Miller. 1973. Integrating timber and wildlife management planning. Forest Chronicle 49: 247- 250. Thompson, W.A., P.H. Pearse, G.C. van Kooten and I. Vertinsky. 1992. Rehabilitating the backlog of unstocked forest lands in British Columbia: A preliminary simulation analysis of alternative strategies. Pages 99-130. In: P.N. Nemetz. Editor. Emerging Issues in Forest Policy. UBC Press, Vancouver, B.C. Turner, M.G. 1989. Landscape ecology: the effect of pattern on processes. Annual Reviews ofEcology and Systematics 20: 17 1-197.  173  Vincent, J.R. and C.S. Binkley. 1993. Efficient multiple use forestry may require land-use specialization. Land Economics 69(4): 370-76. Wallin, D.O., Swanson, F.J., and Marks, B. 1994. Landscape pattern response to changes in pattern generation rules: land-use legacies in forestry. Ecological Applications 4(3): 569-580. Walter, G.R. 1977. Economics of multiple use. Journal ofEnvironmental Management 5: 345-56. Wilcove, D.S. 1985. Nest predation in forest tracts and the decline of migratory songbirds. Ecology 35: 1-22. Wilcove, D.S., C.H. McClellan, and A.P. Dobson. 1986. Habitat fragmentation in the temperate zone. Pages 233- 256. In: M.E. Soule. Editor. Conservation Biology: The Science of Scarcity and Diversity. Sinauer Associates, Sunderland, Massechusetts, USA. Williams.D. 1993. Timber supply in British Columbia: The historical context. Proceedings of a seminar on “Determining Timber Supply and Allowable Cuts in BC”. Association of B.C. Professional Foresters. Vancouver, B.C. 102 p. Worthington, N.P. and G.R. Staebler. 1961. Commercial thinning of Douglas fir in the Pacific NorthWest. USDA Forest Service Technical Bulletin No. 1230. Yahner, R.H. 1988. Changes in wildlife communities near edges. Conservation Biology 13: 133-143. Zhang, D. 1994. Implications of Tenures for Forest Land Value and Management in British Columbia. Unpublished Ph.D thesis. University of British Columbia, Vancouver. 131 p.  (Appendix) 174  APPENDIX List oftables used in Appendix Table 1 Description ofcodes used in the tables andfigures in appendices  176  Table 2 Conversion ofoldgrowth and second growth to modIedforest  178  Table 3 Silvicultural regimes examinedfor Douglas-fir spruce and redcedar  179  Table 4 Prices used in stand level economic analysis  182  Table 5 Real price increase factors used in stand andforest level economic analysis  182  Table 6 Prices used in forest level economic analysis  182  (Appendix) 175  List offigures used in Appendix Figure 1  Map of British Columbia showing the location of the study area; Reveistoke 1  183  Figure 2 Composition and distribution ofspecies groups in Revelstoke 1  184  Figure 3 Douglas fir (51=19): actual and affordable costs for commercial thinning with and withoutfertilization  185  Figure 4 Douglas fir (SI=19):feasibility as indicated by discounted net revenue for precommercial thinning with and withoutfertilization andpruning  185  Figure 5 Douglas fir (SI =19): actual and affordable costs for commercial thinning with and withoutfertilization andpruning  186  Figure 6 Douglas fir (SI =19):feasibility ofcommercial thinning as indicated by discounted net revenue with and withoutfertilization andpruning with real price increases  186  Figure 7 Douglas fir (SI =12): actual and affordable costs for pre-commercial thinning with and without pruning  187  Figure 8 Douglas fir (SI =12): feasibility, as indicated by discounted net revenue, ofpre commercial thinning with and withoutfertilization andpruning  187  Figure 9 Red cedar (SI =13): actual and affordable costs for pre-commercial thinning with and without pruning  188  Figure 10 Red cedar (SI =13) feasibility, as indicated by discounted net revenue, of precommercial thinning with and without pruning  188  Figure 11 Spruce (SI =18): actual and affordable costs for pre-commercial thinning and commercial thinning with and without pruning  189  Figure 12 Spruce (SI =18):feasibility, as indicated by discounted net revenue, ofprecommercial thinning and commercial thinning with and withoutpruning  189  Figure 13 Spruce (SI =10): actual and affordablefor pre-commercial thinning with and without pruning  190  Figure 14 Spruce (SI 10): feasibility, as indicated by discounted net revenue, of pre commercial thinning with and without pruning  190  Figure 15 Revelstoke 1 showing the distribution ofresource emphasis areas under the IU system  191  Figure 16 Revelstoke 1 showing the distribution of timber zone, non-timber zone and reserves under the SUsystem  192  (Appendix)  176  Table 1 Description of additional codes used in the Tables and Figures in Appendix Code  i  Description  C  western redcedar  c13  western redcedar (SI=13)  c21  1  western redcedar (SI=21)  CT  commercial thinning  CT(112)  commercial thinning with removal of1/2 volume ofstand  CT(1/3)  commercial thinning with removal of1/3 volume ofstand  Cw  western redcedar  Dfir  Douglas-fir  F  Douglas-fir  Fl  first application offertilizer  fl2  Douglas-fir (SI=12)  119  Douglas-fir (SI=19)  F2  second application offertilizer  g  good and medium sites  g&m  good and medium sites  H  hemlock  notreat  no silvicultural treatment  p  poor sites  P1  pruning with first lfi’ (3m height)  P2  pruning with second flfl (5m height)  PC12  precommercially thinned to 120 sph  PC5  precommercially thinned to 500 sph  PC8  precommercially thinned to 800 sph  PCT  precommercial thinning  (Appendix)  Table 2 (Continued). See title Description Code PH  pure hemlock  Revi  Revelstoke 1 or Akokolex drainage  S  spruce  slO  Engelmann spruce (SI=1 0)  s18  Engelmann spruce (Sfr=18)  SI @50  site index at breast height age of5O years  sph  stems per hectare  spp  species type  Sw  spruce  177  (Appendix)  178  Table 2 Conversion of old-growth and second-growth to modified forest. Stand Id.  Existing forest type  Converted Forest Stand Type  Site Index  1  fir/larch/pine (g&m)  Douglas-fir_g&m (119)  19  2  fir/larch/pine (p)  Douglas-fir_ p (fl2)  12  3  cedar (g&m)  cedar_g&m (c21)  21  4  cedarQ,)  cedar_p (c13)  13  5  Pure hemlock  Engelmann spruce_g&m (s18)  18  6  hemlock (g&m)  Engelmann spruce_g&m (S 18)  18  7  hemlock (p)  Douglas-fir_p (112)  12  8  spruce (g&m)  Engelmann spruce_g&m (s18)  18  9  spruce (p)  Engelmann spruce_p (slO)  10  10  fir/larch/pine (g&m)-r  Douglas-fir_g&m (119)  19  11  fir/larch/pine (p)- r  Douglas-fir_p (112)  12  12  cedar (g & m)-r  western redcedar_g&m (c21)  21  13  cedar (p)-r  western redcedar_p  14  Pure hemlock-r  Engelmann spruce_g&m (s18)  18  15  hemlock (g&m)-r  Engelmann spruce_g&m (S 18)  18  16  hemlock (p)-r  Douglas-fir_p (112)  12  17  spruce (g&m)-r  Engelmann spruce_g&m (S 18)  18  18  spruce (p)-r  Engelmann spruce_p (slO)  10  (C 13)  13  (Appendix) 179  Table 3. Silvicultural regimes examined for Douglas-fir, Engelmann spruce and redcedar. Species Dfir Dfir Dfir Dfir Dflr Dfir Dfir Dfir Dfir Dfir Dfir Dfir  Dfir Dfir Dfir Dfir Dfir Dfir Dfir  Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dflr Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir  SI  19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19 19  sph 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600  PCT@4m P6m none none none none none none none 800 800 none none 800 none 800 none 1200 800 1200 800 1200 1200 500 1200 500 none 800 800 800 800 800 none 500 500 500 500 500 none none none none none none none 800 none 800 none 800 none 800 1200 none 1200 800 1200 800 1200 500 1200 500 none 800 800 800 800 800 none 500 500 500 500 500 none none none none none none 800 none none 800  P2@lOm F1,F2 none none none none none none none none none none none 800 none 800 none none none none none 800 none none none 500 none none none none none 800 none none none none 500 none 4m none none 4m 4m none 4m none none 4m 4m 800 4m 800 4m none none 4m 4m 800 4m none 4m 500 4m none 4m none 4m 800 4m none 4m none 4m 500 4m, 6m none 4m, 6m none 4m, 6m none 4m, 6m none 4m, 6m none  CT  none 1/3 1/2 1/3 1/2 1/3 1/2 none none none none none none none none none none none none 1/3 1/2 1/3 1/2 1/3 1/2 none none none none none none none none none none none none 1/3 1/2 1/3 1/2  (Appendix) 180  Table 3 (continued). See title Species Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir  Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir Dfir  Dfir Dfir Dfir  SI 19 19 19 19 19 19 19 19 19 19 19 19 19 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12 12  sph 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600  PCT@4m P@6m 800 none 800 none none 1200 800 1200 800 1200 500 1200 500 1200 none 800 800 800 800 800 none 500 500 500 500 500 none 1200 800 1200 800 1200 500 1200 500 1200 none 800 800 800 800 800 none 500 500 500 500 500 none 1200 1200 800 800 1200 500 1200 500 1200 none 800 800 800 800 800 none 500 500 500 500 500 none 1200 800 1200 800 1200 500 1200 500 1200 none 800  CT P2@lOm F1,F2 1/3 4m, 6m 800 1/2 6m 4m, 800 none 4m, 6m none none 4m, 6m none none 4m, 6m 800 none 4m, 6m none none 4m, 6m 500 none 4m, 6rn none none 4m, 6m none none 4m, 6m 800 none 4m, 6m none none none 4m, 6m none 4m, 6m 500 none none none none none none none none 800 none none none none none 500 none none none none none none none none 800 none none none none none none none none 500 none 4m none none none 4m none 4m 800 none 4m none none 4m 500 none 4m none none 4m none none 4m 800 none 4m none none 4m none none 4m 500 none 4m,6m none none 4m,6m none none 4m,6m 800 none 4m,6m. none none 4m,6m 500 none 4m,6m none  (Appendix) 181  Table 3 (continued). See title  spruce spruce spruce spruce spruce spruce spruce spruce spruce spruce spruce spruce spruce  12 12 12 12 12 18 18 18 18 18 18 18 18 18 18 18 18 18  sph 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600  spruce spruce spruce spruce spruce spruce  10 10 10 10 10 10  1600 1600 1600 1600 1600 1600  1200 1200 1200 800 800 800  none yes  cedar cedar cedar cedar cedar cedar cedar cedar cedar cedar cedar cedar cedar  21 21 21 21 21 21 21 21 21 21 21 21 21  1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600 1600  none none none none none none none 1200 1200 1200 800 800 800  none none none  Species Dfir  Dfir Dfir Dfir  Dfir  SI  PCT@4m P@6m 800 800 800 800 none 500 500 500 500 500 none none none none none none none yes yes none none yes yes none none 1200 1200 yes 1200 yes none 800 yes 800 yes 800  yes none yes  yes  yes yes yes yes none yes yes none yes yes  P21Om F1,F2 4m,6m none 4m,6m 800 4m,6m none 4m,6m none 4m,6m 500 none none none none none none none none none none none yes none yes none none none none none yes none none none none none yes  CT none none none none none none 1/3 1/2 1/3 1/2 1/3 1/2 none none none none none none  none none yes none none yes  none none none none none none  none none none none none none  none none none none none yes  none none none none none none none none none none none none none  none 1/3 1/2 1/3 1/2 1/3 1/2 none none none none none none  yes none none yes none none yes  (Appendix) 182  Table 3 (continued). See title Species cedar cedar cedar cedar cedar cedar  sph 1600 1600 1600 1600 1600 1600  SI 13 13 13 13 13 13  PCT@4m 1200 1200 1200 800 800 800  P@6m  CT  P2@lOm F1,F2  none yes yes none yes  none none yes none none  none none none none none  none none none none none  yes  yes  none  none  Table 4 Prices used in stand level economic analysis 10-19cm ) 3 ($/m 80  20-29cm ) 3 ($1m 110  30-39cm ) 3 ($/m 120  40-49cm  Douglasfir  <10cm ) 3 ($/m 30  Redcedar  30  80  85  90  95  Spruce  30  70  100  105  120  Forest stand type  ) 3 ($1m  130  Table 5 Real Price Increase factors used in stand and forest level economic analysis Forest stand type  <10cm (%/p.a)  10-19cm (%/p.a)  20-29cm 30-39cm 40-49cm (%Ip.a) (%/p.a) (%/p.a)  All stand types  0.00  0.00  0.0022  0.0076  0.01  Blend of all classes (Oldgrowth) 0.0014  Table 6 Prices used in forest level economic analysis <10cm  10-19cm  20-29cm 30-39cm 40-49cm  type  ) 3 ($1m  ) 3 ($1m  ) 3 ($1m  ) 3 ($1m  ) 3 ($/m  Blend of all classes (Oldgrowth)  Allstandtypes  30  77  98  105  115  75  Forest stand  (Appendix) 183  Ian:poximoI!Jr 125km  Reveistoke TSA  Revelstoke 1  Figure 2 Map of British Columbia showing the location of the study area, Revelstoke 1.  (Appendix)  184  2500 2000  F(g)  F(p)  C(g)  C(p)  PH  H(g)  H(p)  S(g)  S(p)  F_r(g) F_r(p) C_r(g) C_r(p) PHr H_r(g) H_r(p) S_r(g) S_r(p)  and groups  Figure 3 Composition and distribution of stand groups in Revelstoke 1. (Refer to Table 1 (Appendix page 175) for descrztion ofcodes).  (Appendix)  185  3000  2500  Aff$ Aj  1000  500  0 PC2  PC8  PC_F1  PC_F2 PC8_F1 Treatment Type  PC8_F2  PCT2P2  PCT8_F2P2  Figure 4 Douglas fir (S119): Actual and affordable costs (with real price increase) for commercial thinning with and without fertilization. (Refer to Table 1 (Appendix 1) for description ofcodes).  DNR  Treatment Type  Figure 4. Douglas fir (Site Index 19): Feasibility as indicated by Discounted Net Revenue (with Real Price Increase) for Precommercial thinning with and without Fertilization and Pruning. (Refer to Table 1 (Appendix 1)for description codes).  (Appendix) 186  10000  ..  1  /1  1  3000  ‘  .  0  .  —  2000  —  1000  z V  —4  .  r  0 CT(3)-T  :.•..  CT(2)-T  CT(3)_F1  ...  CT(2)_F1  .  .  .  CT(13)_F2  CT(2)F2  CT(3)_F2P2 CT(?2)_F2P2  Treatment Type  Figure 5. Douglas fir (S119): Actual and Affordable costs (with Real Price Increase) for Commercial thinning with and without Fertilization and Pruning. (Refer to Table 1 (Appendix 1) for description ofcodes).  7000 6000  w 5000 4000.  Li  3000 2000. .  1000 0 CT(V3)-T  CT(1/2)-T  CT(V3)F1  CT(V2LFI  CT(113)F2  CT(112LF2  CT(1’3LF2P2  CT(112)_F2P2  Treatment Type  Figure 6 Douglas fir (S119): Feasibility of Commercial thinning as indicated by Discounted Net Revenue for Commercial thinning with and without Fertilization and Pruning with Real Price Increase. (Refer to Table 1 (Appendix 1)for description of codes).  (Appendix)  187  2000 1800  1600 1400 1200  -  -,  -  -  Act$  U,  1000800 600f 400  PCI2_F2P2  Treatment Type  Figure 7 Douglas fir (Site Index 12): Actual and Affordable costs (with Real Price Increase) for Pre-Commercial thinning with and without Pruning. (Refer to Table 1 (Appendix 1) for description ofcodes).  DNRj 0  Treatment Type  Figure 8 Douglas fir (Site Index 12): Feasibility, as indicated by Discounted Net Revenue (with Real Price Increase), of Pre-Commercial thinning with and without Fertilization and Pruning. (Refer to Table 1 (Appendix 1)for description ofcodes).  (Appendix) 188  1600 1400. 1200 (U  800  8  .  600.  Af  Act$  400 200  0 -20w -400 Treatment Type  Figure 9 Red cedar (SI=13): Actual and Affordable costs (with Real Price Increase) for pre-commercial thinning with and without pruning. (Refer to Table 1 (Appendix 1) for description ofcodes).  0  ID12_P -100  -  (U  -200  -  0 C  I g  -300 -400  -  DNR -  -500-600-  0 CU  0  -700  -  Treatment Type  Figure 10. Red cedar (SI=13) Feasibility, as indicated by discounted net revenue (with real price increase), of precommercial thinning with and without pruning. (Refer to’ Table 1 (Appendix 1) for description of codes).  (Appendix)  189  n’v1  4500 4000 a a 0 a C  3500 ..Aff$  3000  Act$  2500  a  2000  a  1500  /  /  I-  1000 500  PC2  PCS  CT(1’3)  PC8_P2 CT(1’2) Treatment Type  PC_P2  CT(3)_P2  CT(2)_P2  Figure 11 spruce (S118): Actual and Affordable costs (with Real Price Increase) for Pre commercial thinning and Commercial thinning with and without Pruning. (Refer to Table 1 (Appendix 1) for description codes).  3000 2500  a 2000 a C  a  a  z V  a C 0 U  a  CT(13)  CT(112) Treatment Type  Figure 12 spruce (S118): Feasibility, as indicated by Discounted Net Revenue (with Real Price Increase), of Precommercial thinning and Commercial thinning with and without Pruning. (Refer to Table 1 (Appendix 1)for description ofcodes).  (Appendix) 190  0 0 0  _-•_  Afi$  Act$  = 0  E 0  I-.  PCI2  PC12_P2  PC8  PC8_P2  Treatment Type  Figure 13 spruce (SI=1O): Actual and Affordable costs (with real price increase for pre commercial thinning with and without pruning. (Refer to Table 1 (Appendix 1)for description ofcodes).  Treatment Type  Figure 14 spruce (SI=1O): Feasibility, as indicated by Discounted Net Revenue (with Real Price Increase), of Pre-commercial thinning with and without pruning. (Refer to Table 1 (Appendix 1) for description ofcodes).  (Appendix)  191  r I  b P  9 WC3  VPR_WC 1 VPR_WC3  Figure 15 Reveistoke 1 showing the distribution of resource emphasis areas under the IU system. (Refer to Tables 3 (page 50) and 5 (page 52) in textfor description on codes)  (Appendix)  192  1  V  4  Timber zone  Non-timber zone Reserves  Figure 16 Reveistoke 1 showing the distribution of timber zone, non-timber zone and reserves under the SU system.  (Appendix)  193  GLOSSARY OF TERMS Adjacency: an integrated management guideline that specifies that adjoining areas surrounding an harvest cutbiock should not be harvested until the regeneration in the harvest cutbiock reaches specified height or age. Age class: any interval into which the age range of forests or stands is divided for classification and use. The intervals are usually based on age-in-tens. Allowable Annual Cut (AAC): the annual allowable rate of timber harvest from a specified area of land Biodiversity: the diversity seen in living organisms in all its life forms at all levels of organization. It may range from diversity in genetic alleles to diversity in ecosystems. Biogeoclimatic Ecosystem Classification (BEC): a hierarchical classification system with three levels of integration (regional, local and chronological) and combining climatic, vegetation and site factors. a large geographic area with a broadly homogeneous Biogeoclimatic zone: microclimate. Each zone is named after one or more of the dominant climax species of the ecosystem in the zone and a geographic or climax modifier. Clearwood: high value wood that is laid down after pruning or when the bole is clear of living or dead branches (i.e., wood without knots) Diameter at Breast Height (DBH): diameter of the tree at the breast height point which is 1.3 m above ground level. Ecosystem: a functional unit consisting of all living organisms in a given area and all the non-living physical and chemical factors of their environment, linked through nutrient cycling and energy flow. Edge: it is a band on the periphery of a patch of forest that differs abiotically and biotically from the interior and the exterior. Faildown: the amount by which current harvest levels must decline to meet long run sustainable harvest levels. Forest Ecosystem Network (FEN): habitat islands and the system of linkages between them. Forest cover requirements (in British Columbia management context): Specify desired distribution of areas by age or size class groupings in a management unit. They reflect desired conditions for wildlife, watershed protection, visual quality and other integrated resource management objectives. Forest Practices Code (FPC): legislation, standards and field guides that govern forest practices in British Columbia. Forest Renewal Plan (FRP): a major long-term plan, supported by legislation, to renew British Columbia’s forests by improving reforestation, silviculture, cleaning up environmental damage and enhancing community stability and employment within the forest sector. Freegrowing: An established seedling of an acceptable commercial. species that is free from growth inhibiting brush, weed and excessive tree competition.  (Appendix)  194  Green-up period: the time needed for a stand of trees to reach a desired condition (e.g. height) to ensure maintenance of water quality, wildlife habitat, soil stability or aesthetics. Growing stock: the estimated volume of all standing timber, of all ages, at a particular time. Habitat: natural home of plant or animal. Height class: an interval into which the range of tree or stand heights is divided for classification and use. Also, the trees or stands falling into such intervals. Mean Annual Increment (MAI): stand volume divided by stand age. The stand age at which the MAI assumes its maximum value is called culmination age. Harvesting all stands at this age results in a maximum average sustained harvest over the long term. Not satisfactorily restocked (NSR) land: productive forest land that has been denuded and has not been regenerated to the specified or desired free growing standards of the site. Pareto optimality: maximization of welfare of the society. It is the point at which an individual cannot be made better off without making another worse off. Patch: an area of forest that is homogeneous with respect to some attribute, for example, a seral stage. It is a fundamental structural element of the landscape. Polygon: a specific area with definite boundaries that have been authorized for harvest by the Ministry of Forests. Also referred to as harvest unit or cutblock. Protected Areas Strategy (PAS): a process to coordinate all of British Columbia’s protected areas programs and objectives. Protected areas: areas such as federal parks, provincial parks, wilderness areas, ecological reserves and recreation areas that have protected designations according to federal and provincial statutes. Regeneration: the renewal of a tree crop, whether by natural or artificial means. Riparian zones: land adjacent to the normal high water line in a stream, river or lake extending to the portion of land that is influenced by the presence of the ponded or channeled water. Seral: stages in a sequence of biotic communities (the sere) that successively occupy and replace each other in a particular environment over time. Site Index (SI): a measure of the relative productivity of a site, based on height of the dominant trees at an arbitrary age. Snag: a standing dead tree from which leaves and most of the branches have fallen. Sustained yield: a method of forest management that calls for an approximate balance between net growth and amount harvested. Tenure: an interest or right held in Crown land or resource granted by statute (e.g., Forest Act). Timber Supply Area (TSA): an integrated management unit established in accordance with Section 6 of the Forest Act. Visual Quality Objective (VQO): defines a level of acceptable landscape alteration resulting from timber harvesting and other activities. A number of visual quality classes have been defined on the basis of the maximum amount of alteration permitted.  (Appendix)  195  Visual Quality Objective Modification (VQM): alterations may dominate the landscape in this VQO class. Visual Quality Objective Partial Retention (VPR): alterations are visible but not conspicuous. Wilderness Area: an area of land that basically retains its natural character and on -  -  which human impact is transitory and, in the long run, substantially unnoticeable.  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0075321/manifest

Comment

Related Items