UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Phytoecological impacts and management implications of the Douglas-Fir Tussock Moth near Kamloops, British.. Majawa, Andrew Orton 1977-02-16

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
831-UBC_1977_A6 M34.pdf [ 17.16MB ]
Metadata
JSON: 831-1.0075399.json
JSON-LD: 831-1.0075399-ld.json
RDF/XML (Pretty): 831-1.0075399-rdf.xml
RDF/JSON: 831-1.0075399-rdf.json
Turtle: 831-1.0075399-turtle.txt
N-Triples: 831-1.0075399-rdf-ntriples.txt
Original Record: 831-1.0075399-source.json
Full Text
831-1.0075399-fulltext.txt
Citation
831-1.0075399.ris

Full Text

PHYTOECOLOGICAL IMPACTS AND MANAGEMENT IMPLICATIONS OF THE DOUGLAS-FIR TUSSOCK MOTH NEAR KAMLOOPS, BRITISH COLUMBIA by ANDREW ORTON MAJAWA B.S.F. (Hons), University of•British Columbia, 1975 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF FORESTRY in the Department of FORESTRY We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA APRIL, 1977 <£) Andrew Orton Majawa, 1977 In presenting this thesis in partial fulfilment of the requirements f' an advanced degree at the University of British Columbia, I agree tha 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 representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia 2075 Wesbrook Place Vancouver, Canada V6T 1W5 Date 2 7^ AfrJL, (177 ABSTRACT Seven outbreaks of Douglas-fir tussock moth, Ovgyia pseudotsugata McDunnough, have recurred in the interior of British Columbia since 1915. But little is known about their impacts on renewable resources in affected stands. A study wassundertaken to examine effects of the most recent outbreak on understory vegetation and tree productivity near Kamloops, British Columbia. Dry weight forage production was sampled from lmA circular plots under various levels of stand crown cover (0-96%) and density (0-45.9m /ha), as modified by defoliation. Crown cover was determined using a moosehorn, and from vertical photographs obtained with a 160° lens mounted on a conventional camera. Stand density was determined using a 20 factor prism. Increment cores were obtained at breast height, and radial growth analysed under the Addo-X. Ring width behaviour was compared with occurrence of past outbreaks. The ecological literature on 0. pseudotsugata was reviewed. Negligible amounts of forage were obtained from many plots with undefoliated trees. In defoliated plots with live trees, total forage 2 production ranged from 0.0 under 96% crown cover and 45.9 m /ha density 2 to 648.9 kg/ha under 50% crown cover and 16.0 m /ha density. The average yield in small openings was 3667.4 kg/ha. High variability was evident. In one stand, two years following its defoliation and consequent death, total forage yields exceeded those from nearby small openings. Forage yield data were described better by logarithmic models - ii -than by hyperbolic ones, at 95% probability. Impacts on tree growth were not demonstrable one year following defoliation. Many trees recovered even from complete defoliation. Insect outbreaks and periods of slow tree growth coincided, but quite inconsistently. Apparently, most scattered infestation patches develop independently of each other. Grazing values should increase in seriously defoliated stands even without range seeding. On poor sites and in stands managed primarily for forage production, outbreaks of 0. pseudotsugata may be left alone without necessarily endangering remote stands. Selective control favoring better sites managed for tree production should improve efficiency of investing scarce funds in protection of the inventory. Tree growth and insect outbreaks may be under the influence of some regional climatic factor, but local factors are also important. A need remains for long term impact studies on tree growth, forage yield and nutrient status, and other resources. - iii -ACKNOWLEDGEMENTS I thank the committee which supervised me on my graduate program, including this project: Drs. K. Graham and J.H.G. Smith of the Faculty of Forestry; Dr. R.M. Strang of the Faculty of Forestry and Department of Plant Science; and Dr. J.H. Myers of the Department of Plant Science and Institute of Animal Resource Ecology. Dr. Graham was my principal advisor. He and F.A.O. of the United Nations supported me on the program arid project financially. Dr. Smith lent me his camera and moosehorn which I used to estimate stand crown cover; he and Mr. Z. Srejic assisted me in analyses of tree radial growth data. Drs. Graham and Smith encouraged me when the going got tough. Dr. Strang lent me some equipment for vegetation clipping. Through her inquisition of forestry terminology, Dr. Myers necessitated my clarifying some important concepts. I also thank Dr. A. Kozak and Mrs. K. Hejjas of the Faculty of Forestry for invaluable help in statistical modeling of forage yield data on stand characteristics. Dr. R.F. Shepherd of the Pacific Forest Research Center, Victoria, started me on this project. Dr. A. McLean and Mr. L. Haupt, Canada Department of Agriculture, Kamloops, kindly made available laboratory facilities for my use. Mr. R. Chan, Balco Industries Ltd., Kamloops, and Mr. V. Craig, B.C. Forest Service, Kamloops, provided me with useful information. Gunter Schmidt helped me collect most of the data, and kept me company in the field. I am very grateful to all of you. - iv -I should also thank all the I met and talked to in the field, for in the specific study area. loggers, ranchers and graziers telling me some of their experiences - v -TABLE OF CONTENTS Page ABSTRACT i ACKNOWLEDGEMENTS iii TABLE OF CONTENTS v LIST OF TABLES - vi LIST OF FIGURES viiINTRODUCTION . 1 THE INSECT 3. ECOLOGICAL IMPACTS 5 THE STUDY AREA 8 METHODS . . . 16 Understory vegetation . . . . . . .16 Tree growth, survival and salvaging .... 20 RESULTS 24 DISCUSSIONS ~ 69 Impacts .......... 69 Epidemiology ......... 107 Cost of an outbreak ........ 118 Resource use conflicts and pest management strategy . 124 CONCLUSION 130 LITERATURE CITED 132 APPENDIX ' 140 Scientific names of plants ...... 141 - vi -LIST OF TABLES Page Table 1. Ecological conditions prevalent in tree plots where understory vegetation was clipped . 19 Table 2. Extent of sampling for historical radial tree growth, effects of defoliation by Douglas-fir tussock moth on tree growth and understory forage yields ....... 22 Table 3. Extent of sampling for Douglas-fir resilience to, and salvageability of stands following defoliation by Douglas-fir tussock moth 23 Table 4. Average dry weight forage yields in the study 3.X* 6 3, 27 Table 5. Average dry weight biomass of understory vegetation in tree patches in the study area 28 Table 6. Grass yields under various stand stocking and density classes-. . . . . • 30 Table 7. Forb yields under various stand stocking and density classes ....... 32 Table 8. Shrub yields under various stand stocking and density classes ....... 34 Table 9. Total forage yields under various stand stocking and density classes 36 Table 10. Hyperbolic models significant at 95 percent probability level: forage yields ... 41 Table 11. Logarithmic^ models significant at 95 percent probability level: forage yields ... 43 Table 12. Radial growth at breast height in Ponderosa pine and Douglas-fir on a medium site at Mountain View ...... 60 Table 13. A comparison of forage yields from openings in two forest associations in the Interior Douglas-fir zone ........ 72 Table 14. A comparison of forage yields in several plots at Mountain View. ...... 75 - vii -Page Table 15. Percent forage composition under crown cover, by stand density classes ..... 87 Table 16. Average percent change in tree growth at breast height following defoliation in 1975 at Mountain View ....... 91 Table 17. Average percent change in tree growth at breast height by tree size following defoliation in 1975 at Mountain View ..... 92 Table 18. Recovery of completely defoliated trees in a stand at Heffley Creek hillside, and dynamics of their growth ....... 99 Table 19. Recovery of completely defoliated trees on a southeast facing dry site, and dynamics of their growth ........ 100 Table 20. Survival of all completely defoliated Douglas-fir trees tallied in the study .... 100 Table 21. Average regeneration stocking in experimental plots under canopy and in adjacent protected openings ........ 102 Table 22. Stand composition of Douglas-fir trees in defoliated stands before and following salvage logging . . . . . . . 105 Table 23. Distribution of Douglas-fir tussock moth cocoons in lightly and heavily defoliated trees . . 113 - viii -LIST OF FIGURES Page Figure 1. Major outbreaks of Douglas-fir tussock moth in western North America ..... Figure 2. A model of interrelationships between important factors interacting in an outbreak of Douglas-fir tussock moth .... Figure 3. Extent of the outbreak in the study area; sample localities indicated (inside back cover) Figure 4. A fisheye view of a part of Cherry Creek - 1974 Figure 5. Cherry Creek - 1975 ..... Figure 6. Typical Indian Gardens country . Figure 7. A part of Mountain View .... Figure 8. A representative fisheye view of Dairy Creek Figure 9. Heffley Creek - Top: hillside; Bottom: roadside Figure 10. Relationship between grass yields and stand density ....... Figure 11, Relationship between forb yields and stand density ....... Figure 12. Relationship between shrub yields and stand density ....... Figure 13. Relationship between grass yields and stand shocking ....... Figure 14. Relationship between forb yields and stand stocking ....... Figure 15. Relationship between shrub yields and stand stocking ....... Figure 16. Relationship between grass yields and stand shocking ....... Figure 17. Relationship between forb yields and stand - s'.toc - ix -Page Figure 18. Relationship between shrub yields and stand "stocking ........ 53 Figure 19. Relationship between dry weight biomass and stand density ....... 54 Figure 20. Relationship between dry weight biomass and stand.stocking ....... 55 Figure 21. Relationship between dry weight biomass and stahdis|ocking ....... 56 Figure 22. Relationship between total forage yields and stand density ...... 57 Figure 23. Relationship between total forage yields and standtstoeking . . . • • • • 58 Figure 24. Relationship between total forage yields and stand.stocking ....... 59 Figure 25. Behaviour of earlywood width in Ponderosa pine. 61 Figure 26. Behaviour of earlywood width in Douglas-fir . 62 Figure 27. Behaviour of latewood width in Ponderosa pine . 63 Figure 28. Behaviour of latewood width in Douglas-fir . 64 Figure 29. Ring width behaviour in Ponderosa pine . . 65 Figure 30. Ring width behaviour in Douglas-fir ... 66 Figure 31. Behaviour of latewood/Ring in Ponderosa pine . 67 Figure 32. Behaviour of latewood/Ring in Douglas-fir. . 68 Figure 33. Response of understory vegetation at Mountain View ......... 76 , a) Top: Fisheye view - Foreground: Trees ~ defoliated and dead two years ago (1974) *V'* *> ©Background: Negligibly defoliated. . 76 Bottom: Photograph from the interphase between the dead and undefoliated parts of the stand ...... 76 - x -Page Figure 33. b) Fisheye and ordinary views of forage response in one plot in foreground (a, above). Forage yields (kg/ha): Yg = 66.0; Yf = 278.5; Ys = 129.9, under % CC 39m, 40ierls, and basal area of 20.7m2/ha (dead trees) .... 77 c) Zone of transition between the defoliated and undefoliated parts. Note very sparse vegetation in the understory. Bottom: A dense, undefoliated plot near Cherry Creek - 1975. Forage yields (kg/ha): Yt=0.0 Figure 36. Understory vegetation response under a defoliated now dead, open grown Douglas-fir tree at Cherry Creek - 1975. "ayp': Tree characteristics: Diameter = 86.5cm; height = 26.5m; crown width = 14m; ~ . age >150 years. . . . ;'lTo.p^le,ClQ>Ve3cup'iivi'ew^4.nirsuiraner.-u Bottom: Close up view, in winter Note the apparent delineation of the zone of response by crown projection. Downy brome and Pine grass most abundant. 78 d) Understory vegetation in the undefoliated part of the stand. Forage yields (kg/ha): Yg =14.9; Yf = Ys = 0.0, under % CC = 72.^ = lens and basal area 25.3m2/ha. . 78 Figure 34. A conceptual model showing response behaviour •:: b.f'-"forage yields in two hypothetical, defoliated stands with different density levels 81 Figure 35. Top: A dense, completely defoliated plot at Dairy Creek. Forage yields (kg/ha): Y =10.0; Yf = Ys = 0.0, one year following defoliation. Basal area = 2:7.6m2/ha; % CC = 44m> 50iens. Forage yields likely to increase drastically in time if most trees die. .... 83 83 84 85 85 Figure 37. Some defoliated Douglas-fir trees infested with bark beetles. Size preference evident. Note thriving Ponderosa pine . 97 - xi -Figure 38. Lush current year needle crop on small Douglas fir trees, following complete defoliation during the previous year. Evidence of good ' recovery unless redefoliated! Top, middle: Cherry Creek .... Bottom: Indian Gardens .... Figure 39. Abundance of cones on the ground; but no regeneration was evident in some openings. Behind the range/pole is undefoliated plot: Forage yields (kg/ha): Yt = 51.6; % CC = 67m, 64^ens. Cherry Creek, 1974 Figure 40. Structure of residual stands following salvage logging. Marginal tree size Ca. 20cm. .'• Top r ^Zfiey rfoeks public prop rr'tyV Top: Heffley Creek, public property Bottom: Mountain View, private property (Mr. Inskip). Logging progressing into the devastated area. . . . . Figure 41. Shotgun effect: typical nature of Douglas-fir tussock moth infestation on a landscape Figure 42. Active ant colonies ..... Top: Cherry Creek - 1975; colony behind the range pole . . . . Bottom: Mountain View; colony behind the log piece1. r-Nbte rsome de'fbliafioh-;ihqadj aeent - xii -In the past, we have often worked on the premise that all insect and disease outbreaks are detrimental and must be controlled. Perhaps in most instances this has been true, but the effect of the outbreak on wildlife and aesthetic resources may have been a positive factor that needs more recognition. A justification for Federal funds must now be very carefully considered and prepared. The President's Office of Management and Budget is not particularly impressed by the general statement that millions of acres of trees are infested by insects and that a large volume of timber is dying. Nor is it impressed that many livelihoods are tied directly to the forest resource or with similar arguments. What does impress the office are specific benefits that can be derived from the expenditure of alternative amounts of Federal dollars. In other words, the benefit-cost evaluation is directly associated with the proposed project. We must never forget that every dollar appropriated for forest insect and disease protection is one less dollar available for medical research, mass transit, flood relief assistance or similar needs. We must be in a position to point out what effects are specifically associated with forest pests or pollutants when no corrective action is taken. And we must be able to quantify the benefits that will acrue ... under alternative levels of spending for forest pest management and environmental quality evaluation activities. When we have good information, we can make good decisions .... The value of general estimates for analytical purposes is small. We need reliable comprehensive data of forest insect and disease losses (and benefits) that will stand up to close scrutiny. If this information were now available we could evaluate our losses more realistically and use scarce funds and manpower more efficiently .... John R. McGuire, Chief, U.S. Forest Service. At the 1974 Symposium on the Spruce budworm, Alexandria, Virginia. - 1 -INTRODUCTION The latest outbreak of Douglas-fir tussock moth, Orgyia pseudotsugata McDunnough (Lepidoptera: Liparidae), in the interior of British Columbia is the most devastating of all seven recorded outbreaks which have recurred there since about 1915. Outbreaks have occurred quasi-synchronously over a wide geographic region in western North America. They also occur quite regularly with a mode of about eight years between outbreaks in many places in southern British Columbia (Fig. 1). Outbreaks are confined to the interior Douglas-fir"'" forest,,', specifically to the ecotone between lower elevation Ponderosa pine and higher elevation Douglas-fir types. Critical synecological data for zonal ecotones here are lacking (Tisdale and McLean, 1957) even now. Because plants continually contend with each other in the tension zone, there is no stable association in the strict sense of the term. Notwithstanding, an edaphoclimatic "climax" association may be defined here as Douglas-fir-Ponderosa pine-Blue bunch wheatgrass, for purposes of this thesis. No region in the province has more renewable natural resource values and uses converging upon the forest than the southern interior. Critical values include timber, fish, range and forage for wild as well as domestic ungulates, aesthetics, soil and watershed. Inasmuch as various sectors of the public make overlapping demands on the resources within an ecosystem, conflicts exist in their management. Compatibility is possible, but it is not as evident as conflicts. As various "^ Scientific names of plants are available in appendix. CHASC -U i;.:; i:!!! i u.u LlLIOOPT j|j CACMt C^Lfcljcj! SAVCWA !; | NORTH THOMSON KAMLOOPS Hi MONTE LAKk j :. SALMON ARM |; ARMSTROA/G- ! I! VEHNON i [i KkdwA/A ! ii WfSTBANk _T ;:fENT|CTO/M !;: 'OKAM«rAM L-DQ . OLALLA QSOHOQS • !H Ml III i ! I I I ! I II Mill! Ii jliPJ iiilj 01 liillll! ;:CASCAJ)C 7. OKAWQ-AN N.F. • COLV/Ut NF^JM) i:r.:_:::;' i ..^ . • PA LLP as E .j±|q:h-jx|: UMhTlLLP. wr YOSEMlTI A/ F STANISLAUS N.F. /V.F. BOISE /*. |. BOISE COUNTY NEVADA pu.M.BOUbT NF 'VltiEttFR PEAK .P:jx0xt ill |T I P ill -2. /5 '120 ! : « i ! ! 'Jti 4-0 < PPPi a I i mi 1M:H: i:|:i:q; Jxixjxtd: xt TO I ! i.i.U. ttb :px IXE Xtxn: TO 16 * jjxqx XJXXT i ? Mil1 •|-|~HT 1111! Mi ! II 60 PP 11 I i I! I 11 1 Mil m :l±t| rr 75 ESI "TTT Mi I i Pi I LIGHT SEVERE ! i M M | ] S | j I j I Figure 1. Major outbreaks of Douglas-fir tussock moth in western North America. The chart represents outbreaks recorded in numerous accessible literature. Main sources: Sufcden, 1957; Canadian Forestry Service's Forest Insect and Disease Survey reports; U.&.D.A.-U.S.D.I. ... Douglas-fir tussock moth pest management plan, 1973). - 3, -resource values converge on an ecosystem, so do their managers and users. Since 1973, when the current outbreak surfaced, concern has been expressed about its impact on the ecosystem. Arguments about the impacts are mainly empirical, and are compartmentalized often with groups promoting polarized views. Some graziers and wildlife managers contend that defoliation is beneficial because it results in increased understory forage yield and quality. Some forest managers, on the other hand, maintain that notwithstanding the low site quality of infested stands, the wood, fiber and watershed values exceed those of other sympatric resources. The problem of resource use becomes more complex as various public interests become involved. Many arguments are based on intuition and empirical evidence. Specific data are lacking. As some arguments are speculative, they may provide questionable ground on which resource management decisions are based. The need for relevant data is obvious. This thesis investigates and reports some real and potential ecological impacts of Douglas-fir tussock moth outbreaks in a part of the Kamloops Forest District, in the British Columbia interior. The data should aid resource managers there in making well-founded decisions regarding outbreaks. The thesis also examines some ecological aspects of the insect, and discusses a rational approach towards a strategy for its management. The insect The Douglas-fir tussock moth is a native of western North America where it is one of the most destructive forest defoliators. It was described by McDunnough (1921) from a holotype or specimen of - 4 -several paratypes from Chase, B.C. It was he who separated the insect from Hemerocampa vetusta gulosa complex. In the American literature, the insect is still sometimes referred to as Hemerocampa pseudotsugata (Grant et al., 1975; Harwood, 1975). In Canada, the genus Orgyia appears to have been completely accepted since about 1961. The life cycle of the insect varies along ecoclines within its wide geographic habitat. Adults emerge, mate and lay eggs within a short period in summer. The eggs, in diapause, remain unhatched throughout the winter. Larvae emerge in spring and begin feeding in the upper and outer parts of tree crowns. The larvae trek to those parts immediately following emergence. The apparent photopositive reaction is possibly triggered by hunger (tension) as in Eastern spruce budworm, Choristpneura fwniferana (Wellington, 1948). In the European Orgyia antiqua the photopositive reaction is inhibited by some tactile sensors in the forelegs: at the end of a branch, absence of tactile stimuli leads larvae to revert to exploratory maneuver (Zanforlin, 1970) so that the insect does not fall off. Five larval instars may have one less - are most common. Although as many as seven instars have been mentioned (U.S.D.A., 1973b), this has not been ascertained 2 in the scientific literature. Sexual dimorphism of non-sexual characteristics is exhibited by adult Douglas-fir tussock mo ths: tf<? have normal lepidopterous wings, and nonfunctional vestigial ones. 2 It is well known that rearing insects on low quality food may increase the number of larval moults (Leonard, 1970). To what extent the endocrine system - corpora allata^>' e'edysia! Jglands - is influenced by the food is not clear. - 5 -Host "preference" also varies by region. In B.C., Douglas-fir is the preferred host; sympatric trees such as Ponderosa pine are rarely attacked. The preference for Douglas-fir was implied by McDunnough (1921) in his classical paper where he reported some rearing results. The preference is evident also from Forest Insect and Disease Survey records, and field observations. In the U.S. Pacific Northwest, White fir and Grand fir are preferred, but Douglas-fir is often attacked (Eaton and Struble, 1957). Other hosts there and farther south include Western larch, Western hemlock, Subalpine fir and Engelmann spruce (Balch, 1932; U.S.D.A., 1973b). The insect also feeds on lesser vegetation especially when tree foliage is depleted. Beckwith (1976) points out that whether the variation in host preference reflects biological races of the insect is not known. Ecological impacts Ecological impacts can be evaluated by assessment at three stages in the sequence of events: before, during and following application of the ecological force. Potential impacts are evaluated before the activity. In the U.S., this is often done as a necessary part of environmental project feasibility analysis under the 1960 Federal National Environmental Policy Act Section 102. The resulting Environmental Impact Statements attempt to predict outcomes of socioecological significance. The U.S. Forest Service has already undertaken impact studies for a few forest insects including the Douglas-fir tussock moth. The voluminous 1973 Draft Environmental Impact Statement for the tussock moth (U.S.D.A., 1973b) in the Pacific Northwest - 6 -centered on whether and how the insect should be controlled. A benefit-cost ratio of 13 (D.A. Graham, 1974) clearly justified control. However, it was difficult to decide on how to control the insect because, as Harwood (1975) pointed out, after a survey of the literature Stark found very few "scientific papers" on this insect pest. Impacts and their magnitudes vary with intensity, frequency, severity and timing of the respective ecological force. Impacts are also a function of the ecosystem prevalent when the force is applied. Figure 2 illustrates possible relationships between many factors involved in determining the impact of an outbreak of Douglas-fir tussock moth. The model becomes more complex when a management decision is superimposed on the system. Some of the relationships are hypothetical, but they should be appreciated by decision makers and resource managers involved in the problem. The tussock moth directly affects.various parts of the vegetation. This, in turn, affects the stand of which secondary vegetation may be a major component. The extent to which impacts are evident at various levels of stand, forest etc. in the hierarchy depends mainly on the spatial or geographic extent of an infestation. Reciprocal impacts are real: I infer from.data presented by Condrashoff and Grant (1962) that depletion of tree foliage results in a change of preference for oviposition and diapause to parts of trees nearer to the ground, where predation may be intense. Weather, elevation and other external factors also influence impacts of defoliation. As the model indicates, there is an impact at any level in the hierarchy. For a small outbreak, a pest management decision may - 7 -Management decision: Action/No-.- action Weather, site factors ilevation\ soil ) aspect Tree spp.: composition Lesser veg.: composition spatial arrangement yield growth conditiosurvival quality reproduction availability FOREST RESOURCES ^^oil-watershe^^ GRAND ECOLOGICAL' IMPACT Figure 2. A model of interrelationships between important factors interacting in an outbreak of Douglas-fir tussock moth. Arrows indicate flow of influence; thickness of arrows indicates relative intensity or importance of influence. NOTE reciprocal impacts e.g. Trees—>Douglas-fir tussock moth. - 8 -probably be based on impacts at the stand level. For extensive outbreaks, however, it is necessary that decisions be made following analyses of impacts on resources at the regional level. Such analyses should provide the rationale or raison d'etre and its basis for the fundamental decision of control or no-control. Until recently most ecological impact analyses were pre occupied with evaluating effects of (chemical) control decisions, often bypassing the direct ecological impacts of the insect itself. It was often assumed a priori that damage was serious enqugh to warrant some control. Furthermore, emphasis was on investigations of effects on the animal community, mainly birds, fish and ungulates. Insects were examined in followup studies by entomologists to determine efficacy of the control measures. Apparently, the long time it takes for impacts to show in some parts of the ecosystem discouraged serious research there. Thus, impacts on the plant community have largely been ignored in Douglas-fir tussock moth outbreaks. In this thesis I emphasize impacts on the phytocoenose - the tree and lesser vegetation. Even within this restriction, for practical reasons, the need for evaluating only the more important impacts is evident. Other impacts are not ignored, however. The study area The study area is in what is generally referred to as the North Thompson and Kamloops in Forest Insect and Disease Survey reports. It is MominWtedl-by unevenaged, mostly second growth Douglas-fir-Ponderosa pine stands. - 9 -The interior is a part of the Dry Forest Biotic Area which is "faunistically most closely related to, and indeed forms a northern extension of the Great Basin Complex" (Munro and McT. Cowan, 1947). No detailed synecological studies relevant for the outbreak zonal ecotones are available in the accessible literature. Broadly, the area is a part of Rowe's (1972) M-l or Ponderosa pine Douglas-fir section of the Montane Forest Region, and Krajina's (1959, 1965) Interior Douglas-fir Biogeoclimatic Zone, Dry or Pinegrass subzone. . It approximates Beal's (1974) Pseudotsuga Agropyron Spicatum association - "a topoedaphic climax since both topography and soil are necessary for its establishment" - of the wetter Southern Cariboo Zone. In the Similkameen valley, the broad Pseudotsuga menziesii zone (McLean, 1969) embraces ecological conditions common in the study area. The ecotone does not exactly fit in any of Brayshaw's (1965) distinct associations: It lies between his Pinus ponderosa - Agropyron Spicatum var. inerme and Pseudotsuga menziesii - Agropyron spicatum associations'. I consider the zonal ecotone as Pseudotsuga menziesii - Pinus ponderosae -Agropyron spicatum quasi-association, which represents a dynamic situation and is not truly climax. Study plots were restricted to the southern part of the North Thompson Valley near Dairy.and Heffley.creeks, and Mountain View, and to the south of Kamloops Lake near Cherry Creek and Indian Gardens (Figure 3 - folded inside back cover). <Cdiamagrostis,rubescens was a minor part of the vegetation in a few plots. Other lesser vegetation included Balsam root, Timber milk-vetch, JunegrassSagebrush, Needle-and-thread, and Kentucky bluegrass. Most of these plants are - 10 -invaders on overgrazed sites. . In most, plots, soils were generally deep, fine to medium textured. A description of each stand follows. 3 Cherry Creek - 1974: This stand was severely defoliated in 1974. Plots were located between 653 and 720 m elevation, and between 3 and 25 percent slope. Northeast facing slopes are dominant. The soil is. deep, sandy loam. The forest floor depth averaged 3.3cm; in a few plots it was as deep as 7.6cm. Two years following defoliation, twenty three percent of trees tallied had been or was infested with bark beetles. The smallest tree infested had a breast height diameter of 7.9cm. Only one Ponderosa pine tree was infested. Cattle and horses were grazing in the area at the time of sampling. Plots for understory vegetation were, however, located where visual evidence indicated no grazing activity. (Figure 4). Cherry Creek - 1975: This stand was defoliated in 1975. Plots were, located between 683 and 720m, and between 3 and 25 percent elevation and slope respectively. It is generally similar to Cherry Creek - 1974. (Figure 5). Information of years when experimental stands were defoliated was given by Dr. R.F. Shepherd, Canadian Forestry Service, Victoria, B.C. Figure 5. Cherry Creek - 1975. - 12 -Indian Gardens: This is a dry site. The soil is loamy clay with a few rock outcrops scattered throughout. The duff layer averaged 2.5cm in depth. The stand approximates a pure Douglas-fir type as no Ponderosa pine trees were tallied in the sample plots. Plots were within 827 -921m and 15 to 25 percent ranges of elevation and slope respectively. (Figure 6). Figure 6. Typical Indian Gardens country. - 13 -Mountain View: One stand sampled here was defoliated in 1975. The soil is loamy clay. Plots were located between 518 and 636m, and between 5 and 20 percent elevation and slope respectively. The stand is on a southeast facing slope. Some plots were located in another stand which was defoliated in 1974. Virtually all trees were dead at time of sampling, two years later. A slope of 35 percent was quite common; elevation ranged from 647 and 671m. (Figure 7). Figure 7. A part of Mountain View. - 14 -Dairy Creek: The stand is characterized by deep loamy clay soil, and a forest floor averaging 6.4cm deep. In many plots the forest floor was severely cracked, indicating extreme dryness. Part of the stand was fenced off from grazing for the second consecutive year. The stand faces south. Elevation and slopes of plots were between 624 and 878m, and 5 and 20 percent respectively. (Figure 8). Figure 8. A representative fisheye view of Dairy Creek. To face page 15 Figure 9. Heffley Creek. Top: hillside Bottom: roadside - 16 -Heffley Creek: Two stands both along the creek were sampled here. One stand on the north side of the creek was only lightly defoliated in 1975. Elevation of 588m and slope of 20 percent were typical. The other stand is on the south side on a very steep hillside. I refer to these stands as Heffley Creek - roadside and hillside. The soil is very deep sandy loam. On the hillside, a steep slope of 88 percent makes the soil prone to mass wasting; deep ravines are common here. Probably the angle of repose is not much less than the prevailing 42°. Depth of the forest floor averaged 7.6cm. Bunchgrass is exceptionally dense here. The tussock moth swept through this stand in 1975 and left virtually all Douglas-fir trees completely defoliated. (Figure 9). Methods Understory vegetation In the literature it is evident that forage (= herbage) yield on forested range land is a function of interacting ecological factors. In this thesis I studied the influence, of stand density and stocking as modified by defoliation, on understory forage productivity. In each location, temporary plots were chosen in a maximum of six groups of trees representing different ecological conditions:-No defoliation - high density, low density; Partial defoliation - high density, low density; Complete defoliation - high density, low density. Defoliation classes here refer to the patch rather than individual trees, and the terms high and low density are relative. - 17 -For each plot and adjacent area in the stand, I noted stand type, aspect, slope, elevation, soils and the forest association. The purpose was to define part of the ecological domain within which the data would be valid. All plots reported here were within the ranges of 517 to 921m and 3 to 88 percent for elevation and slope. From each plot center, a sweep was made with a prism of basal area factor 20, to obtain the number of "in" trees. The patch density was determined using principles of variable plot sampling (Dilworth and Bell, 1972). I tallied all "in" trees taller than 1.4m or breast height and recorded the species, breast height diameter (dbh), height, whether it was live, dead or undetermined, recovery potential if defoliated, and presence or absence of secondary insects. For every crown I recorded the following: width, total length, length of dead portion or top kill, and of partially defoliated, and undefoliated sections. The proportion of current year foliage was estimated ocularly. This was possible because of clear distinction in color and position of current and older foliage. 2 In the past, the use of 9.6 ft microplots was traditional among American range ecologists. Its origin dates back to 1949 when it was proposed independently by Frischknecht and Plummer, and Campbell and Cassady. The rationale was that biomass from the plot in grams is equivalent to one tenth of yield in pounds expected from one acre, or 2 43560 ft . Later, Canadians adopted the same plot size so they could quickly judge the capability of their range land by comparing with 2 American data. Currently, lm plots are used with increasing frequency 2 in forage productivity studies. Grams of biomass from a lm plot are - 18 -equivalent to one tenth of yield in conventional kilograms-per-hectare units. 2 Within each tree plot I systematically established four-lm microplots." Circular plots have the smallest perimeter per unit area (Van Dyne e_t al_. , 1963), and this minimizes edge effects which may bias results in forage yield studies. I used a metal loop to determine microplot boundary: Tree plot center Tree plot with imaginary, irregularly shaped boundary enclosing all "in" trees. microplot Microplots were established also in openings. In each microplot a field assistant and I clipped all lesser vegetation - shrubs, forb.s, grasses - rooted within the plot, at root collar. Tree seedlings were counted but not clipped. The three components of understory vegetation were separated in the field, put in bags and later stored in a cold room. The vegetation was then dried at 50°C for 50 hours, and weighed to the nearest one hundredth of a gram - 19 -In the laboratory. We did this part of the field work during the summer of 1976. We estimated crown cover under each plot in two ways. One reading was obtained from above the center of each microplot using a moosehorn invented by Robinson (1947) and later described by Garrison"-(1949). For reliable and precise moosehorn crown cover estimates, as many as forty readings per one quarter acre plot are recommended (Robinson, 1947; Bonnor, 1968). Our readings fall within this acceptable limit. In spite of difficulties involved in holding the moosehorn vertically and steadily, especially in adverse weather conditions in the field, we found the technique quite precise (Table 1). Table 1 Ecological conditions prevalent in tree plots where understory vegetation was clipped. Data for crown cover are based.on 172 microplots, and for basal area on 36 tree plots. __ , _ -Percent crown cover Basal area (m /ha) moosehorn wide angle lens Mean 50. 10 50. ,90 21. 09 Minimum 0. 00 0. ,00 0. 00 Maximum 96. 00 90. ,00 45. 90 Stand, deviat. 28. 80 25, ,70 12. 20 Coef. of Var. 0. 58 0. .51 0. 58 - 20 -The second technique involved the use of hemispherical or' "fisheye" (160°) lens mounted on a conventional.Pentax camera (Brown and Worley, 1965; Bonnor, 1967), loaded with a high speed color slide film. I took crown photographs of each tree plot from ground level, and determined crown closure from prints.using a dot grid. This technique requires clear skies, calm weather and some familiarity with photography. It becomes very expensive if the field should be revisited to take more photographs so spoiled ones may be replaced. But it provides a semi permanent record of the plots. The risk of a film developer losing or even mixing up good photographs is always present. Therefore, plots should be marked, albeit temporarily, and films developed as quickly as possible so that plots can be revisited quickly without difficulty if required. The use of a fixed angle lens in this technique gives crowns of taller trees a better chance of being included in a photograph. Yet it is recognized that shorter trees may have a shading influence of their own. This probably biases estimates of crown cover in uneven aged stands, especially if the angle is narrow and tree plots large. In this study, most tree crowns appeared in the photographs, as in each patch variation in tree height was small. Tree growth, survival and salvaging. To investigate possible association between occurrence of outbreaks and trends in tree radial growth, and effects of the tussock moth on tree survival, we obtained increment cores from trees in each tree plot. We cored two Douglas-fir trees - one tree of average dbh, the other of maximum dbh, and two Ponderosa pine trees both of average - 21 -dbh. One of the Ponderosa pine trees was. from inside, and the other outside the plot. The number of increment cores obtained in each plot was variable as some plots did not have any Ponderosa pine trees, and some trees outside one plot also served for another adjacent plot. Increment cores were preserved in plastic straws; later I analysed them for earlywood, latewood width using the Swedish Tree Ring Machine, the Addo-X. Only two of 229 cores were discarded because of serious defect of rot at a pitch pocket. The freezing technique (Francis et al., 1972) used to detect rings, in rot pockets is valid for studies involving only ring counts for age and site index determination. The technique is inappropriate for studying. ring behaviour ..because in situ moisture may cause changes in cell width. For each Ponderosa pine tree outside the plot, we recorded similar parameters as we did for each tree inside the plot. Because of apparent high variability in tree growth, I suspected that in.order to detect reliable impacts on tree radial growth, some intensive sampling of increment cores within one area was necessary. So in addition to the above, I cored 40 trees for ring width behaviour analyses on a medium site at Mountain View. I did this after tree growth had ceased in winter of 1976. I cored two trees from each of the following 20 treatments: Diameter classes (cm): <^15.0, 15.1-25.0, 25.1-35.0, >35.0; and for each diameter class the following crown conditions: Ponderosa pine control; Douglas-fir defoliation classes: control, 5-25%, 30-60%, >65%. All the cored trees had been defoliated in 1975; they were live at. time of sampling. The purpose was to examine historical correlations between Table 2. Extent of sampling for historical radial tree growth, effects of defoliation by Douglas-fir tussock moth on tree growth and understory forage yields. Site quality - B.C. Forest Service: M - medium, L - Low, P - poor. Elevation % Site Tree Trees tallied Trees cored Cores Lesser veg. (m) Slope quality plots "::t, ZF analysed plots (incl. Tot. %F F Py open) Dairy.') iGreek 624-878- 19-26 Cherry Cr. - 1974 653-'7.20 3-25 - 1975 683-778 3-35 M-P P-L P-L 72 57 67 97 93 88 12 12 12 8 7 27 32 34 28 28 28 Heffley rCr.- Roadside 558 - Hillside 610 20 88 M M-P 2 3 13 100 100 12 16 Mountain View - 1974 617-671 35 M-P 2 20 100 - 1975 519-636 5-20 M 7 89 96 12 31 8 32 Growth response 561-702 Indian Gardens 872-921 5-20 15-26 M P 40 50 100 32 8 8 2 80 18 20 36 408 91 37 229 172 Table 3. Extent of sampling for Douglas-fir resilience to, and salvage-ability of stands following defoliation by Douglas-fir tussock moth. Elevation On) % Slope No, Strip Size(m) Trees F tallied Py Trees F dead Py Avge Basi (m2/V Heffley Cr. Hillside 610 88 1 8x31 48 15 (unsalvaged) 610 80 2 12x40 121 3 21 — — Heffley Cr. 824 5 1 9x214 44 3 2 (Balco-Salvage Oper.) 800 10 2 8x92 12 2 2 - -800 10 3 8x122 79 1 31 - -790 25 4 31x61 28 13 6 - -(adj. stands unlogged) Prism plot 790 20 1 8 - 1 - 37 793 25 2 14 - 6 - 32 793 25 3 12 - 2 - 28 793 10 4 5 5 - 1 23 793 32 5 4 4 — — 41 - 24 -past outbreaks and tree growth during corresponding periods, and tree growth responses to different intensities of defoliation in 1975. All increment cores were obtained at breast height. Table 2 shows the extent of sampling undertaken in this part of the project. Remarkable resilience of Douglas-fir following defoliation by the tussock moth in western U.S. was implied in the literature by Caroline and Coulter -"(19.75). My field observations suggested that mortality of infested trees varied mainly with size and degree of defoliation. Therefore tallies were taken of individual trees in 2 representative strips at Heffley Creek - hillside, where all trees had been completely defoliated in 1975. The strips measured 8 x 31m and 12 x 40m. A total of 172 trees were tallied in both strips. For each tree, we recorded the species, condition, dbh, height, crown width and length, recovery potential (subjectively determined) and absence or presence of secondary insects. Strips were also located in areas where salvage logging had been undertaken. Trees from there were also used in the resilience study. This part of the study, done in summer of 1976, also served for investigating the nature and impact of salvaging infested stands. (Table 3). Results The efficiency of a sampling system is often estimated as the reciprocal of the sample variance. The efficiency of system "i" relative to "j" is given by o2 / a2 ->. S2 / S2 . when-costs of sampling per miWre notu%qualr, efficlincy^is^modifIbB to'1/|txCV^](Freese, 1962) • where C -r sMtpling-' c'o'sfc 'per"utt-irt,CV^=: coefficient of variation. In this study, one film exposure cost on the average $0.15 for purchase, $0.12 for developing into transparencies, and $0.77 for developing into prints for dot grid analysis for a total of $1.04. The 2 sample unit under consideration is one lm microplot where understory vegetation was clipped. Whereas one moosehorn reading was required for each microplot, four microplots fell within each tree plot whose crown cover was photographed. Thus each sample unit cost $0.26. Stocking estimates by the moosehorn and wide angle lens are compared in Table 1. The relative efficiency of the moosehorn system: RE , = (C x CV2) lens moosehorn 2 (C x CV ) moosehorn = 26C x 0.512 1C x 0.582 = 20.10 Although it produces a higher sample variance, the moosehorn is more efficient than the photographic system by more than twenty times. I have ignored capital costs of the equipment, incidentals of mailing films and slides and riskg. of losing a film, slides or prints. The relative efficiency of the moosehorn calculated above is therefore a conservative estimate. - 26 -Basal area and percent crown cover may be biologically related, but correlation analysis showed the two variables to be statistically independent (r = 0.26). The wide range of basal area under similar levels of stocking or crown cover (Smith, 1974) and vice versa, is indicative of the complex dynamics of stand development. The low correlation between stocking and density is not unexpected because each varies with aspect, slope, elevation, soils and other factors which are unaccounted for here. In the first 46 microplots sampled, forage was not separated into grass, forbs and shrubs because of limited manpower. However, it was possible later to isolate the vegetation from 126 plots. In developing regressions for total biomass I used the data from all 172 plots, while for regressions describing grass, forbs and shrubs I used data from 126 plots. This means that regression coefficients in models of the three components are not additive to those of total biomass. Tables 4, 5, 6, 7, 8, 9 summarize the yield data from all plots. Evidently, variation in yield decreases in the order of grass, forbs and shrubs, and with increasing stocking and density. [Note that coefficients of variation are not percentages]. In subsequent statistical analyses variables are defined as follows: Independent variables: 2 : Basal area (m /ha) X£ : Percent crown cover by moosehorn X : Percent crown cover by wide angle lens - 27 -Dependent variables: Y : Total forage yield Y : Grass yield g Y^ : Forb yield Y : Shrub yield s All yields are dry weight (Kg/ha) as described in methods. Table 4 Average dry weight forage yields in the study area. Data are based on 126 microplots, except for total which are based on 172 micropolots. Grass Yields (Kg/ha) Forbs Shrubs Total Mean Minimum Maximum Stand, dev. Coef. of Var. 88.5 0.0 743.0 139.3 1.57 47.8 0.0 599.0 108.6 2.28 48.6 159.8 0.0 0.0 534.5 1175.4 97.7 216.3 2.01 1.35 Table 5. Average Dry weight biomass (kg/ha) of understory vegetation in tree patches in the study area. Each moosehorn and yield datum is based on 4 microplots. Basal area % Crown Cover Grass Forbs Shrubs Total (m2/ha) moosehorn lens Cherry Cr. 1974 0.0 13.8 16.1 18.4 25.3 27.5 29.8 Cherry Cr. 1975 0.0 16.1 20.7 23.0 23.0 29.8 41.3 Indian Gardens 0.0 16.1 23.0 29.8 46.0 0 0 -25 39 -63 63 -28 69 -67 64 -68 63 -41 55 -0 0 48.9 49 64 -40 42 -20 36 -60 66 -70 40 37.6 40 51 -0 0 260.6 53 57 306.8 70 62 6.1 70 53 80.8 70 50 0.0 - - 207.8 - - 112.0 - - 51.6 - - 126.5 - - 0.0 - - 24.7 - - 250.1 126.5 178.9 354.3 - - 84.9 - - 60.6 - - 80.0 - - 50.6 3.6 12.5 53.7 - - 125.5 32.3 252.5 545.4 19.6 30.6 357.0 0.0 31.4 37.5 4.5 0.0 85.3 0.0 0.0 0.0 Table 5 continued. Basal area % Crown Cover ' Grass Forbs Shrubs Total (m2/ha) moosehorn lens Dairy Cr. 0 .0 0 0. 417 .3 344 .3 142, .7 904 .3 18 .4 68 74 12 .6 0 .0 0. .0 12 .6 23 .4 75 75 303 .5 8 .1 9 .7 321 .3 25 .3 61 45 90 .5 21 .8 0. .0 112 .3 27 .5 44 50 10 .0 0 .0 0 .0 10 .0 34 .4 69 68 6 .8 0 .0 0 .0 6 .8 36 .7 70 60 38 .7 0 .0 0 .0 38 .7 Mountain View 0 .0 0 0 102 .2 262 .1 120 .5 484 .8 13 .8 61 62 53 .9 57 .4 9 .0 120 .3 23 .0 60 71 3 .9. 83 .0 28 .1 115 .0 23 .0 61 49 28 .8 1 .5 0 .0 30 .3 23 .0 88 78 15 .7 0 .8 19 .8 36 .3 39 .8 91 90 2 .3 0 .5 8 .1 10 .9 41 .3 45 55 44 .5 8 .6 0 .0 53 .1 Dead '74 20 .4 39 40 66 .0 278 .5 129 .9 474 .4 25 .3 72 72 14 .9 0 .0 0 .0 14 .9 Heffley Cr. - Road 0 .0 0 0 62 .7 86 .9 272 .4 422 .0 13 .8 76 75 19 .0 13 .7 60 .5 93 .2 16 .1 83 67 4 .7 0 .0 0 .0 4 .7 Hill 0 .0 0 0 554 .4 61 .6 132 .2 748 .2 20 .7 52 57 104 .9 10 .5 0 .0 115 .4 20 .7 64 58 48 .2 10 .7 68 .0 126 .9 20 .7 71 75 77 .8 0 .0 2 .0 79 .8 Table;.6. Grass yields (kg/ha) under various stand stocking and density. Crown cover data are from moosehorn % CC class 1: 0-10; 2: 11-30; 3: 31-70; 4: 71-96. Basal area (m2/ha) 0.0 - 11.5 11.6 - 23.0 23.1 - 34.4 34.5 - 46.0 Crown cover class 1 >2 .3 '4 :1 •. /2: I, 3 k 4 •. l; " 2 3 3 4 4 .1 1.1 2 3 4 30.7 13.6 6.3 6.4 0.5 21.9 35.7 25.8 0.0 0.0 76.6 526.4 11.5 23.3 83.5 0.0 0.0 2.7 234.5 0.0 80.5 9.3 63.4 13.5 88.6 217.6- 380.4 119.1 181.7 57.9 20.4 67.7 248.5 388.5 124.1 37.7 0.0 0.0 36.5 169.1 0.0 76.7 0.0 0.0 0.0 193.6 276.0 72.1 42.0 0.0 0.0 0.0 364.6 0.0 42.9 28.5 6.6 0.0 9.0 263.0 5.7 44.8 0.0 43.1 . 0.0 221.1 0.0 14.1 11.1 44.1 360.3 45.1 15.7 16.4 108.2 412.4 0.0 22.1 10.6 272.6 0.0 11.0 10.0 623.7 15.5 4.0 0.0 124.0 0.0 57.9 180.0 0.0 0.0 Table 6 continued Basal area (m2/ha) 0.0 - 11.5 11.6 - 23.0 Crown cover class 123412 3 4 53.2 136.0 18.7 51.5 34.9 0.0 75.9 0.0 0.0 25.0 112.6 29.2 48.9 85.4 91.0 101.0 52.2 8.1 621.4 8.2 743.0 5.7 206.9 197.0 396.0 6.0 41.4 175.0 85.5 77.0 1.0 128.7 83.0 23.1 - 34.4 34.5 - 46.0 12 3 4 12 3 4 Mean 215.92 St.dev. 205.59 Coef Var. 0.95 90.13 55.35 116.00 109.4 1.29 1.98 40.30 44.18 42.44 58.21 1.05 1.32 24.87 5.36 37.71 8.01 1.08 1.49 Table 7. Forb yields (kg/ha) under various stand stocking and density. Crown cover data are from moosehorn. % CC class 1: 0-10; 2: 11-30; 3: 31-70; 4: 71-96. Basal area (m2/ha) 0.0 - 11.5 11.6 - 23.0 23.1 - 34.4 34.5 - 46.00 Crown cover class '•; 1 23412 3 4 12 3 4 12 3 4 254.8 499.7 0.0 0.0 0.0 0.9 0.0 34.0 0.0 0.0 18.5 0.0 0.0 0.0 0.0 0.0 0.0 197.8 0.0 0.0 0.0 13.6 0.0 0.0 178.7 78.4 0.0 12.6 0.0 0.0 0.0 110.5 0.0 32.4 0.0 17.3 0.0 0.0 18.7 0.0 21.7 54.4 0.0 0.0 1.8 0.0 0.0 0.0 20.3 0.0 0.0 0.0 52.4 0.0 0.0 0.0 0.0 0.0 0.0 40.5 0.0 16.9 0.0 0.0 0.0 36.2 0.0 0.0 0.0 0.0 578.6 0.0 0.0 0.0 0.2 74.4 124.6 2.3 0.0 599.0 146.0 1.0 0.0 125.3 39.5 2.2 0.0 304.3 5.8 33.4 421.7 0.0 0.0 267.9 15.3 0.0 54.5 56.0 0.0 225.0 141.2 0.0 Table 7 continued Basal area (m2/ha) 0.0 - 11.5 11.6 - 23.0 23.1 - 34.4 34.5 - 46.00 Crown cover class , 123412 3 4123 412 3 4 25.2 130.2 0.0 37.5 300.7 0.0 60.0 183.2 0.0 61.1 18.9 76.0 0.2 1.1 0.0 108.1 0.0 41.7 0.4 0.0 0.0 42.1 0.0 0.0 Mean 151.07 St.dev. 167.17 Coef.Var. 1.11 40.13 5.00 70.61 10.66 1.76 2.13 6.30 3.43 15.13 6.88 2.40 2.01 0.02 0.23 0.06 0.64 3.00 2.78 Table 8. Shrub yields (kg/ha) under various stand stocking and density. Crown cover data are from moosehorn. %CC class 1: 0-10; 2: 11-30; 3: 31-70; 4: 71-95. Basal area (m2/ha) 0.0 - 11.5 11.6 - 23.0 23.1 - 34.4 34.5 - 46.0 Crown cover class 1 2 3 4 1.2:3 44 i.l .2 3 3 , 4 .... 1 : 2 3 4 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 131.8 122.5 125.7 0.0 27.5 0.0 0.0 147.5 0.0 0.0 0.0 22.4 0.0 0.0 178.5 0.0 6.3 0.0 0.0 0.0 0.0 58.5 0.0 0.0 0.0 0.0 0.0 0.0 331.4 32.3 18.6 0.0 0.0 0.0 34.3 0.0 0.0 0.0 0.0 0.0 0.0 0.0 192.4 0.0 0.0 0.0 0.0 0.0 9.0 534.5 0.0 36.1 0.0 0.0 0.0 283.2 0.0 20.8 0.0 0.0 236.5 0.0 27.2 0.0 0.0 117.3 0.0 13.0 0.0 0.0 9.4 18.0 0.0 217.1 34.4 121.8 0.0 35.9 0.0 0.0 0.0 0.0 0.0 165.4 0.0 0.0 280.5 0.0 0.0 Table 8 continued Basal area (m2/ha) 0.0 - 11.5 11.6 - 23.0 23.1 - 34.4 34.5 - 46.0 Crown cover class 1234123412341234 141.0 0.0 0.0 393.8 306.1 12.9 321.0 153.2 0.0 233.9 60.1 0.0 34.9 18.2 136.9 102.1 249.0 0.0 108.1 0.0 0.0 0.0 0.0 259.1 0.0 0.0 7.9 Mean 174.20 St. Dev. 133.97 Coef Var. 0.77 33.49 18.20 74.80 33.77 2.23 1.97 0.0 5.54 undefined .,11.08 2.00 0.0 5.41 undefined 12.09 2.24 Tab)le 9. Total forage yields (kg/ha) under various stand stocking and density. Crown cover data are from moosehorn. %CC class 1: 0-10; 2: 11-30; 3: 31-70; 4: 71-95. Basal area (m2/ha) 0.0 - 11.5 11.6 - 23.0 23.1 - 34.4 34.5 - 46.0 Crown cover 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 186.5 111.3 163.5 124.6 57.6 0.5 235.0 34.3 124.5 182.5 0.0 132.6 59.2 132.4 147.0 111.4 298.5 20.5 59.8 89.5 0.0 285.5 181.5 85.4 38.6 6.4 147.4 23.6 105.5 0.0 226.7 126.2 59.5 137.2 323.2 0.0 0.0 13.5 348.0 69.7 50.5 0.0 20.5 0.0 63.4 20.4 445.8 99.4 33.4 386.7 0.0 0.0 57.9 0.0 326.7 59.3 98.9 420.9 22.8 35.7 0.0 34.3 386.6 20.0 11.6 40.3 23.3 111.0 0.0 0.0 193.6 513.2 66.9 72.1 80.5 45.3 0.0 9.0 609.4 60.7 42.9 131.7 181.7 0.0 838.4 100.6 97.8 124.1 55.0 0.0 540.5 80.0 34.9 131.1 0.0 44.1 1175.4 86.5 42.9 62.3 0.0 108.4 604.1 32.5 37.4 28.5 6.6 871.6 3.4 30.0 0.0 43.1 966.1 6.3 128.0 11.1 464.2 648.9 91.3 16.4 601.7 234.5 0.0 10.6 Table 9 continued ... Basal area (m2/ha) 0.0 - 11.5 11.6 - 23.0 23.1 - 34.4 34.5 - 48.0 Crown cover class 12341 2 3 412 3 412 3 4 .486.5 296.0 18.7 '386.5 248.5 0.0 441.9 201.4 0.0 444.0 276.0 .42.1 407.0 0.0 91.0 394.9 5.7 8.1 717.4 0.0 955.9 45.71i 457.0 124.6 612.2 155.4 89.4 5.8 0.0 151.3 90.9 141.2 336.1 43.9 128.7 Table 9 continued Basal area (m2/ha) 0.0 - 11.5 11.6 - 23.0 23.1 - 34.4 34.5 - 48.0 Crown cover class 12341 2 3 412 3 412 3 4 90.9 548.9 539.3 295.5 45.3 108.0 197.7 6.0 83.1 175.4 85.5 Mean 518.08 St. Dev. 256.63 Coef Var. 0.50 117.33 141.02 133.33 79.07 61.37 146.04 145.55 108.35 0.52 1.04 1.09 1.37 83.85 37.12 92.15 50.10 8.58 100.81 50.05 44.15 58.37 12.22 1.20 1.35 0.50 1.17 1.42 - 39 -Scattergrams (Figures 10 to 24 inclusive) indicated non linear relationships between the dependent and independent variables. Therefore, strict linear models were not examined for regression modeling. Constants in X + 1 and Y + 1 in hyperbolic and logarithmic analyses respectively are used to avoid mathematical problems because in some cases X and Y equaled zero. The logarithmic transformation makes variances along the regression line more uniform. For the logarithmic model, the standard errors of the estimate were transformed into nonlogarithmic form. The transformation procedure is as follows: The SEE = \ | SS /df r r N where SS^ = sum of squares residual df = degrees of freedom residual r Transformation: SS n E j=l (7j + 1) - 10 (b0 + pi Xij) .th , . j 2 = j forage yield n = number of observations Y. = Y , ; f individually J j ' gJ ' j ' Sj X_^ = i1**1 independent variable; i = 1, 2 here bQ, b^ = regression coefficients defined formally. - 40 -= n-m-1 m = number of independent variables in the significant model under consideration. 1/2 n (bQ + Z b± Xi;j) tn-m-1] (y + 1) - 10 x The models reported below (Tables 10, 11) were significant at 95 percent probability level. Analyses were by least squares technique - elimination procedure. Some dependent variables are described by two significant models; the models are in the order they were encountered through elimination. The second model is shown for checking how much precision is lost by dropping one significant variable. This becomes relevant when it may be costly to measure a variable. Coefficients of determination are noftfan'sf drifted "to iriOnldgarithmic form: therefore it would be unwise to compare fitness of hyperbolic with 2 logarithmic models using R values. It is acceptable to compare between logarithmic models using these values. Better fit of logarithmic models is evident. Yields are also shown graphically in figures 10 — 24 inclusive. Enlarging plot size by combining data from 4 microplots (Table 5) did not reduce standard errors significantly. This resulted in an increase of standard errors in some logarithmic models. Apparently, - 41 -Table 10. Hyperbolic models probability level: significant at forage yields. 95 percent Y = b0+b1x1+b2(i/x1+i) SEE 2 %R F Y t = 141.356-2 .9378X-J+365 . 387 (l/X^+l) 147.212 54.238 100.150 : 71.7137+444.595 (1/X^l) 148.468 53.178 193.097 Y g = 47.9099+170.009 (1/X +1) 123.035 22.633 36.275 Yf = 15.1861+136.377 (1/X^+l) 95.1174 23.952 -.39.056 Y s = 9.0021+165.614 (1/X +1) .'73.6212 43.673 96.143 Y = b +b1X0+b. (l/X.+l) o 1 z z z Y t = 169.248-1.4151X2+345.092 (1/X2+1) 146.974 54.386 100.748 : 81.902+164.971 (1/X2+1) 148.661 53.060 192.135 Y g = 52.4075+164.584 (1/X2+1) 123.304 22.295 35.578 Yf = 144.7689-1.8557X 92.9141 27.430 46.881 Y s = 13.2240+161.056 (1/X2+1) 73.7918 43.410 95.126 Y = b +b1X.+b„ (1/X.+1) + b„X„+b. (l/X.+l) oil/ 1 J Z 4 Z Y t = 167.764+348.384 (1/X +1)-1.5260X2 146.269 54.820 102.538 • 71.7132+444.595 (1/X +1) 148.468 53.178 193.079 Y g = 47.9099+170.009 (1/Xj+l) 123.035 22.633 36.275 Yf = 259.417-5. 9193X^-3363.590 (1/X +1) +3255.490 (1/X2+1) 92.359 29.454 16.979 • 18.5076+133.330 (1/X2+1) 95.048 24.060 39.295 Y s = 9.0091+165.614 (1/Xj+l) 73.621 43.673 96.143 - 42 -Table 10 continued... Y = b +b.X,+b. (1/X +1) o 1 3 2 J SEE 2 %R F Y t = 200.798-1.8972X3+314.084 (1/X3+1) 147.617 53.986 99.138 Y g = 52.3948+164.643 (1/X3+1) 123.293 22.308 35.605 Yf = 144.550-1.9323X 94.103 25.567 42.529 Y s 13.216+161.093 (1/X3+1) 73.783 43.426 95.181 Y b +b1X1+b0(l/X1+l)+b_X_+b(1/X.+1) Y t = 304.978-3.6461X -2.0709X +209.885 (1/X +1) 145.208 55.739 70.521 170.698-3.4118X +344.718 (1/X3+1) 146.959 54.395 100.788 Y g = 213.932-3.1847X -1.1880X3 121.524 25.131 20.643 190.743-4.9396XX 123.509 22.037 35.047 Yf = 144.550-1.9323X 94.103 25.567 42.592 Y s = 9.0021+165.614 (1/X +1) 73.621 43.673 96.145 - 43 -Table 11. Logarithmic^Q models significant at 95 percent probability level: forage yields. log(Y+l) = = b +b,Xn o 11 SEE 2 %R F log(Yt+l) = 2.5699-0.0423X1 130.590 34.410 89. 189 log(Y +1) = 2.0582-0.0349X1 44.104 27.171 46. 262 log(Yf+l) = 1.6524-0.0443X1 :16.505 39.514 81. 007 log(Ys+l) = 1.6258-0.0452X1 13.373 38.477 77. 550 log(Y+l) = o 12 log(Yt+l) = 2.5206-0.0167X2 124.434 29.930 72. 610 log(Y +1) = 2.0173-0.0130X2 41.084 20.364 31. 709 log(Yf+l) = 1.7835-0.0201X2 22.399 43.455 95. 293 log(Ys+l) = 1.5435-0.0163X2 10.810 26.955 43. 745 log(Y+l) = VblXl+b2X2 log(Yt+l) = 2.7357-0.0289X1-0.0089X2 200.815 39.428 55. 003 : 2.5699-0.0423X1 145.987 34.410 89. 189 log(Y +1) = 2.0582-0.0349X1 44.108 27.171 46. 262 log(Yf+l) = 1.8897-0.0228X^0.0131X2 29.945 48.619 58. 195 • 1.7835-0.0208X2 22.398 43.455 95. 293 log(Ys+l) = 1.6258-0.0452X1 13.375 38.477 77. 550 - 44 -Table 11 continued... log(Y-KL) = b +b,X3 o 1 . . SEE 2 %R F log(Yt+l) = 2.6169-0.0185X3 150.326 28.968 69.328 log(Y +1) o = 2.1062-0.0154X3 50.601 24.320 39.842 log(Yf+l) = 1.7829-0.0209X 22.409 40.560 86.614 log(Yg+l) = 1.6382-0.0189X 13.662 31.089 55.941 log(Y+l) b +b1X1+b„X. o 11 2 3 log(Yt+l) = 2. 7749-0.0296X --0.0093X3 221.696 38.655 53.245 2.5699-0.0423XX 130.590 34.407 89.174 log(Y +1) = 2.2138-0.0232X1--0.0080X3 66.179 30.605 27.124 • 2.0582-0.0350X1 44.104 27.171 46.262 log(Yf+l) = 1.9013-0.0255X -•0.0128X3 30.869 47.446 55.522 : 1.7829-0.0209X 22.409 40.560 84.614 log(Ys+l) = 1.7902-0.0327X -•0.0084X3 21.353 41.725 44.034 1.6258-0.0452X 13.373 38.477 55.473 - 46 -FIG. 1 1 . RELATIONSHIP BETWEEN FORB YIELDS PND 5TPND DENSITY Y=(49.916 * 0.903*)-! 45.0 0.0 S.D -1 10.0 X1 —1-*— 15.0 20.0 25.0 30.0 STAND BRSPL PREP (SQ.M/HR) 35.0 40.0 50 - 47 -F1G.1 2 RELATIONSHIP BETWEEN SHRUB YIELDS AND STAND DENSITY Y=(42.247 K 0.901)-1. or JZ X a i Oo $ CO £° CO in cn-T 10.0 -r1 20.0 25.0 STAND 0RSAL AREA 30.0 ISQ.M/HAJ 15.lT -1 50.0 - 48 -- 49 -- 50 -3* FIG.l 5. RELRTI0N5HIP BETWEEN SHRUB YIELDS AND STOCKING Y=134.954 * 0.963X)-1, 40.0 ;CROWN COVER 60.0 MOOSEHORN 00.0 ~l— 90.0 ~1 100.0 52 -FIG. 17- RELATIONSHIP BETWEEN FORB YIELDS AND STOCKING Oo t— X CD Q I CD Oo Y=(60.660 X 0.953 20.0 30.0 40.0 50.0 60.0 70.0 X CROWN COVER - WIDE-ANGLE LENS 80.0 -M—tjl so.o 100.0 - 53 -1G. 18. RELATIONSHIP BETWEEN SHRUB YIELDS RND STOCKING Yr(43.471 X 0-.957 J-l , 20.0 I 30.0 40.0 50.0 J CROWN COVER - W1DE-60.0 ANGLE LENS 70.0 BO.O 90.0 100.0 - 54 -- 55 -o FIG. 20- RELATIONSHIP-BETWEEN DRY WT. BIOMASS AND STOCKING GR: Y=U04.064 X 0.97] 1-1. FO: Y=(60.744 X 0.955 )-]. SH: Y=(34.954 X 0.963 )-]. 40.0 50.0 60.0 XCR0WN COVER - MOOSEHORN T 70.0 BO.O i 90.0 100.0 - 56 -. RELATIONSHIP BETWEEN DRV WT. BIOMASS AND STOCKING GR: Y=U27.703 X 0.965X)-1. FO: Y=C60.660 X 0.953X)-]. SH: Y=(43.471 X 0.957X)-]. GRASS FQRB5 SHRUBS FIG. 23. RELATIONSHIP BETWEEN TOTAL FORAGE YIELDS AND STOCKING 1 Y=(331.589 X 0.962X)-]. 0.0 10.0 20.0 30.0 40.0 50.0 60.0 70.0 60.0 90.0 100 ;CROWN COVER - MOOSEHORN - 59 -Table 12. Radial growth (.01mm) at breast height ln Ponderosa pine and Douglas-fir on a medium site at Mountain view. Trees were defoliated in 1975; increment cores were.extracted, in winter 1976. Each average is based on 16 cores. Lw: latewood; Ew: earlywood; R: ";otal annual ring. Diameter class (cm) Averages . v , *i5 15.1-25.0 25.1-35.0 >35' Deg.of Yr.of - x . • defol: growth Lw . Ew R." Lw/R Lw Ew R Lw/R Lw Ew R Lw/R Lw Ew R Lw/R Lw Ew R Lw/R Py 0 1975 9.8 33.0 42.8 .228 18.8 69.8 88.5 .212 14.3 46.8. 61.0 .234 20.2 80.0 100.8 .206 15.9 57.4 73.3 .217 1976 20.3 56.0 76.3 .266 . 22.3 64.0 86.3 .258 15.0 50.0. 65.0 .231 38.0 98.5 136.5 .278 23.9 67.1 91.0 .262 DF 0 1975 30.5 66.5 .97.0 .314 32.3 87.3 119.5 .270 27.3 98.5 125.3 .217 17.5 74.5 92.0 .190 26.9 81.7 108.6 .248 1976 12.3 38.8 50.5 .243 6.8 14.5 21.3 .318 33.0 52.3. 85.3 .387 8.5 23.5 32.0 .266 15.1 32.1 47.3 .320 L 1975 9.5 25.8 35.3 .270 18.5 46.8 65.3 .284 15.3' 44.8 60.0 .254 13.3 52.0 65.3 .203 14.1 42.3 56.4 .250 1976 6.0 11.5 17.5 .343 14.0 42.5 56.5 .248 14.3 27.8 42.0 .339 11.3 43.3 54.5 .206 11.4 31.3 42.6 .267 M 1975 20.3 57.8 78.0 .200 15.3 .45.8 61.0 .250 38.8 104.5 143.3 .271 47.0 80.8 127.8 .368 30.3 72.6 102.5 .290 1976 19.8 54.3 74.0 .267 . 21.5 78.3 99.8. .216 17.8 83.5 101.3 .175 29.0 95.0 124.0 .234 22.0 77.8 99.8 .221 H 1975 36.5 80.5 117.0 .312 31.5 88.3 119.8 .263 .. 16.0 57.0 73.0 . .219 9.3 35.3. 44.5 .208 23.3 65.3 88.6 .263 1976 22.0 57.8 79.8 .276 18.5 88.0 106.5 .174 10.0 35.5 45.5 .220 6.3 22.3 28.5 .219 14.2 50.9 65.1 /218 03 - 69 -more precision may be obtained by increasing the number of sample plots, instead of plot size. Data on radial growth in trees in 1975 and following that year's defoliation are summarized in table 12. Figures 25 — 32 inclusive represent historical radial growth at breast height in trees from the whole study area during the past century. The Douglas-fir curves are based on 150, and the Ponderosa pine curves on 79 increment cores. Discussion Impacts Understory forage yields and their variability decrease with increasing stand basal area and percent crown cover. The general trends evident in the results presented here are common in the literature, but apparently, mathematical, models best describing forage yields vary. Dodd (1969) found a good lit of simple linear models between percent crown cover and forage yields in "undisturbed" stands in higher elevation P. menziesii - oalamagrostis association a few miles northwest of Kamloops. He admitted that curves would describe the data better. Nevertheless he applied linear models "for being more useful (sic p. 51)." Later Dodd e_t al. (1972) fit both linear and semi-logarithmic, models on the same data and found both models significant. But because the SEE are in different units, it is not clear which form of model describes the data better. - 70 -A high linear correlation (r = 0.985) between forage yields and stocking in Ponderosa pine stands in eastern Washington was reported by McConnell and Smith (1970). Pase (1958), Jameson (1967) and others have found nonlinear relationships between yields and stand density and stocking. Gaines e_t al_. (1954) reported a second degree parabola with 2 "k" positive (linear: y = bQ + b^x + b^x ) as describing forage yields on stand density: but they cautioned that the upturn in the curve might have been due to inadequacy of data. On the other hand, they correctly alluded to the possibility that the upturn might have signified a point beyond which stand density was increasing at the expense of stocking, thus making more space and light available for forage production. Jameson (1967) cited a third degree or cubic parabola^ ("linear: 2 3 y = bQ + b^x + b^x + b^x ) as giving a best fit of his forage yields on stand density and stocking. A biological hypothesis for the shape of the continuous function is difficult to conceive. These data were obtained under in situ or undisturbed crown cover conditions. In the Southern Pine Region, herbaceous growth is also inversely related to stand density and crown closure; and forage, yields are highest in openings (Blair and Brunett, 1976). In my analyses, standard errors of the hyperbolic functions are much higher than those of logarithmic functions. Both types of models are significant. They describe the data adequately. Because of high errors associated with them, the hyperbolic models would give very wide confidence intervals if used for prediction. The logarithmic models are invariably better fitting. The coefficients of determination are misleading here because of differences in units of the dependent - 71 -variable in the two kinds of models. In their work in Montana Foothills Bunchg'rass Ranges, Van Dyne et al. (1963) used coefficients of variation to conclude that grasses and forbs were less variable than shrubs. In general, Muegler (1976) concluded similarly. In my study area, grasses are most variable, and forbs more variable than shrubs. Basal area and crown cover are required to predict total forage yields well. But separate predictions of grass, forbs and shrubs require only basal area. •The importance of basal area is not unexpected, as it is indicative of site quality and biological capacity. In most of the multiple regression models coefficients of determination and SEE indicate that crown cover can be ignored in predicting forage yields without necessarily losing much precision. This does not mean crown cover is not important in determining forage yields; it means basal area is more important. Table 13 shows forage yields for two forest associations within the Interior Douglas-fir Zone. Only yields from openings are compared because of lack of stand density data in Dodd (1969), and because of possible interaction between density and stocking in this study. Williams Lake is farther north, cooler and more moist than Pass Lake. As my data are from a driest area of the three, they fall within reasonable expected ranges. At most, 54 percent of the variation in total forage yields was due to basal area and percent crown cover. These independent variables accounted for even smaller amounts of the variation in separate yields of grass, forbs and shrubs. Likely, other factors are important in determining forage production in this ecotone. Gains et al. (1954) showed a very high correlation between grass and forb - 72 -Table 13 A comparison of forage yields (Kg/ha) from openings in two forest associations in the Interior Douglas-fir Zone. Data for Pass and Williams Lakes extracted and further analysed from Dodd (1969). Data for the other association are from six localities studied here. Pseudotsuga - Calamagrostis Pseudotsuga - P.Ponderosa -Pass Lake (Dry site) Williams Lake Agropyron spp. N. Thompson Valley - So. of Kamloops Lake 632.8 459.4 932.5 1538.8 Avge(s:.: 674.9 207.8 354.3 545.4 904.3 484.7 422.0 Avge.:486.4 yields and forest litter amounts. A thick mor forest floor may Imother vegetatively regenerating plants; it makes mineral soil inaccessible to seeds. Severson and Kranz (1976) reported very poor logarithmic fit of forage yield data on basal area in aspen stands. While the yields decreased as proportion of Ponderosa pine increased, predictability increased. They employed a double sampling technique of ratio estimation to obtain dry weight from fresh weight of forage, instead of drying all samples. Therefore in their regression analysis - 73 -they actually used forage yield estimates. It is not indicated, whether the error associated with the use of estimates was accounted for. It is not known whether the insignificance of their models was associated with the double sampling procedure. Notwithstanding, the authors are of the opinion that their models, were insignificant because they did not tally the most important independent variable. Probably in Aspen and other vegetatively regenerating stands, "roots, total biomass or growth" are more predictive of forage yields than basal area or crown cover. Increases of understory vegetation yields in response to tree defoliation by Douglas-fir tussock moth are evident. Greatest increases should be expected in cooler and more moist microsites in the ecotone, nearer the Pseudotsuga calamagrostis association and along creeks. On dry sites, where yields are ordinarily low, forage produc tion may increase several fold, but it does not amount to much except possibly where complete defoliation results in mortality of large groups of trees. In their exploratory work, Tisdale and McLean (1957) observed forage yields under several canopies and basal area levels in several seres in the interior. They noted a pioneer aspen sere with 2 basal area of 20.7 m /ha yielded an average of 270 kg/ha of forage, and 2 the climax Douglas-fir sere with a basal area of 52.1 m /ha yielded only 114 kg/ha. Forage yields in transition seres of Lodgepole pine and mixed conifer/aspen were not given. In a pine stand which was killed by D&ndroctonus beetles, probably D. ponderosae (monticolae)3 forage yields were "... 50 percent greater than for comparable site occupied by abnormal stand of Pinus." This appears to be the first - 74 -time an.insect was recognized and documented in the literature as an important ecological force in range management in the interior. Munro and McT. Cowan (1947) had recognized logging and fire in this regard. Transmissivity of insolation, and basal area in Ponderosa pine stands are significantly related in a hyperbolic fashion (Solomon et al., 1976). Reducing stand density provides more light to the understory. Following defoliation, more moisture and light are available under the canopy; and animals frequent dead patches thereby effecting soil scarification. In one stand, two years following its death, forage yields exceeded those in openings (Table 14; Figure 33). This also illustrates the importance of moving shade provided by the dead trees or snags. In adjacent undefoliated stands, forage yields were extremely low. Not only does defoliation provide more light and moisture to understory vegetation, it also makes available more nutrients which leach out of the forest floor and frass. When defolia tion results in tree mortality, competition - for these resources is greatly reduced and understory vegetation flourishes. Dolph (1973) gave an empirical estimate of about 7,900 animal unit months as the magnitude of range benefits following a tussock moth outbreak in the U.S. Pacific Northwest. From his data, it is not clear whether these benefits are from 70.5 thousand or 80.thousand hectares. Defoliation may have different effects on forage quality and yield. Increases in yield do not necessarily imply increased capacity for range land. In a black tupelo forest in the southern U.S. extensive defoliation by the forest tent caterpillar, Matacosomaid%sstv£ay-'•reduced wild game, population. This was reflected in hunters' success. Besides its Table 14. A comparison of forage yields (kg/ha) in several plots at Mountain View. The two stands merge into each other on the same side of a main haul road, have the same slope, aspect etc. Note higher average yields (534) under dead trees and crown cover 50%; cf. average yields (485) in openings. Dead Stand, Defoliated Undefoliated Stand Openings 2 years before Basal area (m2/ha) 25.3 25.3 % Cr. cover 20 32 50 55 56 64 84 84 Grass 13.6 112.6 85.4 52.2 0.0 10.0 6.6 43.1 124.0 180.0 53.2 51.5 Forbs 499.7 130.2 300.7 183.2 0.0 0.0 0.0 .0.0 304.3 421.7 267.9 54.5 Shrubs 0.0 306.1 153.2 60.1 0.0 0.0 0.0 0.0 35.9 0.0 165.4 280.5 Total 513.3 548.9 539.3 295.5 0.0 10.0 6.6 43.1 464.2 601.7 486.5 386.5 To face page 76 Figure 33. Response of understory vegetation at Mountain View. a) Top: Fisheye view - Foreground: Trees defoliated and dead two years ago (1974). Bottom: Pho,tmgry.^h'. f ^Background: Negligibly defoliated. Bottom: Photograph from the interphase between the dead and undefoliated parts of the stand. To face page 77 Figure 33. Response of understory vegetation at Mountain View. b) Fisheye and ordinary views of forage response in one plot in foreground (a, above). Forage yields (kg/ha): Yg = 66.0; Y = 129.9, under % CC 39m, 40 lens, and basal area of 20.7m2/ha (dead trees). - 77 -To face page 78 Figure 33. Response of understory vegetation at Mountain View. c) T6p.<:^ Zone of transition between, the defoliated and undefoliated parts. Note very sparse vegetation in the understory. d) Bottom: Understory vegetation in the undefoliated part of the stand. Forage yields (kg/ha): Y =14.9; Y=Y =0.0, under % CC = 72.0m = g f s lens and basal area 25.3m2/ha. - 79 -adverse effects on the trees, defoliation encouraged growth of less desirable brush and vines. Apparently, this caused wildlife to abandon the devastated stands for safer ones with overstory (R.C. Morris, 1976). In the Douglas-fir tussock moth situation, the patchy nature of infestations makes it possible for wildlife to find refuge in the general area of an outbreak. Because of possible recovery by some defoliated trees, the question of stand development following defoliation is obviously important in our attempts to determine magnitudes of the range benefits accruing following defoliation.. How much defoliation, and at what frequency can a Douglas-fir tree continue growing or living? What is the recovery potential of the defoliated tree? It appears we must reduce crown cover to less than fifty percent in order to realise geometric increases in forage yield. Eddlemann and McLean (1969) also define fifty percent crown cover as a critical level in Ponderosa pine stands. How long can the yield increasesWe maintained? In an ecological study of fire disturbance in the southern U.S., Blair and Brunett (1976) concluded that kind, intensity and length of intervals (frequency) between disturbance are the most important factors in determining stand composition. This may be true for chronically infested Douglas-fir stands in the interior. Let us suppose we are dealing with a stand with seventy percent crown cover. Crown cover must be reduced to fifty percent to realize significant increased forage-, yields. Also suppose crown cover is forty-five percent following defoliation. While we needed to reduce crown cover by more than twenty percent, all the stand requires - 80 -now is a five percent - probably two years - gain of foliage to revert to its condition of low understory forage yields. Often we are dealing with stands uneven in age, height, stocking and health. When these variables are superimposed on the crown cover in our example above, it becomes obvious that we cannot justifiably generalise about impacts of the defoliation on forage beyond the statements made here. Moreover, the dynamics of stand development are complex (Smith, 1974) enough without superimposing upon them an extra ecological force. Figure 34 is a conceptual model of forage response behaviour in two susceptible stands. Stand a}y has higher density than stand, b,' -for simplicity, assume other factors are equal. The difference in density is reflected in higher in situ forage yields in stand b , Y. > Y Various degrees of defoliation increase forage yields to bo ao ° different levels. Assume equal degrees of defoliation in both stands: ."defoliation results in higher forage yields in stand b, Y, . > Y .. At bx ax very low degrees of defoliation, the difference in productivity between the two stands narrows, (Y. r-. - Y c) < (Y, _ - Y _). Following defolia-bo ao; bl al tion, the tendency is for the stand to revert to its original state: yield approaches the original level, Y, .—^Y, , Y .—*Y . Probably • ° bx ' bo ax' ao both stands require about the same period to revert to the original state of forage productivity; and following a low degree of defoliation, that state would be reached in a shorter period of.time than following higher degrees of defoliations, ta^'pi^t\:)^' tb6<< ^2 * Cbl' anC* t « t < t Clearly, increased defoliation benefits grazing in 3D Q./. 3.1. at least two explicit ways: by increasing yields more or less immediately, and by sustaining the yields for a longer period. —I 1 1 1 1 1 1 20 30 40 50 60 75 90 Degree of Defoliation (%) Figure 34. A conceptual model showing response behaviour of forage yields in 2 defoliated stands with different density levels. Stand a; Stand b. BAA>BAB; YFE > Y3Q - 82 -Low forage yields in, and inaccessibility of thick tree patches (Fig. 35) to domestic and wild ungulates which are of concern in this ecotone (McLean e_t al. , 1970) , leads one to question whether such patches should be included in range land inventory. Criss-crossing by dead trees fallen after their defoliation may impede accessibility, but only to a smaller extent. Only patches occupied by dead trees and trees seriously defoliated to the extent which requires many years to regain suppressive crown cover represent a realistic increase in the inventory. In southern Alabama, Gaines e_t al. (1954) mentioned individual Longleef pine trees between 18 and 36 cm dbh. influencing grass production within a maximum distance of only 2.4 meters from the trunk - crown- widths were not given. On the other hand a group of trees, presumably of similar size, influences grass production within a broader zone of 9 m from the forest edge. The difference is probably due to microclimatic influence of the forest "wall". By corollary, following defoliation, more forage may be realized from along forest walls than under individual open grown trees. Figure 36 shows an impressive response of understory forage under a dead, more than 150 year-old open grown Douglas-fir tree of 86.5cm dbh and 14m crown width, near Cherry Creek. Clearly the zone of response was delineated by the crown. Such response reflects also a greater supply of nutrients from frass and needles following defoliation. Within a stand, increasing crown cover results in reduced forage yields; but it favors grass production at the expense of forbs and shrubs. This is true for most levels of stand density. The same To face page 83 Figure 35. Top: A dense, completely defoliated plot at Dairy Creek. Forage yields (kg/ha): Y = 10.0; Yf=Y =0.0, one year following defoliation. Basal area = 2 27.6m /ha; % CC = 44m, 50 lens. Forage yields likely to increase.drastically in time if most trees die. Bottom: A dense, undefoliated plot near Cherry Creek -1975. Forage yields (kg/ha) Yfc = 0.0 - 83 -To face page 84 Figure 36. Understory vegetation response under a defoliated, now dead, open grown Douglas-fir tree at Cherry Creek -1975. a. Tree Characteristics: Diameter = 86.5 cm; height = 26.5 m; crown width = 14 m; age 150 years. - 84 -To face page 85 Figure 36. b. Top: Close up view, in summer. Bottom: Close up view, in winter. Note the apparent delineation of the zone of response by crown projection. Downy brome and Pine grass most abundant. A - 86 -conclusion is evident from examining forage composition under1 similar stocking levels on increasing density; Stand density and stocking, and probably other factors have an interacting, effect on forage yield and composition (Table 15). At lower stocking and density levels grass is less dominant, but it still constitutes a higher porportion of biomass than forbs and shrubs. In many cases reduction in tree growth following defoliation is considered so common that it is often assumed and used to justify insect control measures. The loss in tree production is relevant only in surviving trees; losses due to mortality may be higher. Tree form may be impared by top kill and formation of spikes, bayonets and forks. This is significant in young stands. In 1973 extensive top kill in a forty one hectare stand near Osoyoos was noted in a Forest Insect and Disease Survey Annual Report. Defoliation may reduce photosynthetic surface enough to affect tree growth. Theoretically, serious reduction in tree growth should be expected because the tussock moth defoliates from above, where current year needles and their biomass are concentrated (Silver, 1962; Smith, 1970). Gordon (1962) studied competitive effects of common understory species on tree radial growth in the "east side pine type" in California. Removal of bunchgrass resulted in increased radial.growth in widely scattered Ponderosa and Jeffrey pine trees. Possibly, in my study area, increased productivity of understory forage following defoliation may prevent residual trees from realising significant growth increment. In his analysis of 4 dominant and codominant coastal Douglas-fir trees, Silver (1962).found the current Table 15. Percent forage composition under crown cover, by stand density classes. Crown cover classes: 1: 0-10% (mostly openings); 2: 11-30%; 3: 31-70%; 4: >71%. Maximum CC = 95% for moosehorn, 90% for wide angle lens. - indicates no plots obtained in treatment. Data from 172 plots. Basal Area Class ( 2 m /ha) 0 - 11.5 11.6 - 23 .0 23.1 - 34. 4 34.5 - 50.0 Crown cover class 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 Moosehorn Grass 40 _ 3 51 49 100 87 83 - - 96 92 Forbs 28 - - - - 97 27 8 0 13 7 - - <1 4 Shrubs 32 - - - 0 22 43 0 0 10 - - 4 4 Wide angle lens - ^ -Grass 40 _ - 49 67 - 84 100 - - 91 21 Forbs 38 _ - 30 17 - 11 0 - - 9 4 Shrubs 32 — _ — - — 21 16 — — 5 0 _ _ 0 75 - 88 -year foliage in the upper 1/3 amounted to over 12 percent more than in the lower 2/3. The current year foliage is preferred food by the tussock moth (Beckwith, 1976).. Further analysis of Silver's data shows there are more needles per linear 2.5 cm (1 inch) of foliated twigs in the lower 2/3: Tree number 12 3 4 No. of needles per linear 2.5cm. Top 1/3: 261 230 229 257 Bottom 2/3: 290 161 299 273 This implies more dilution of current year foliage in the lower parts of the crown; a higher concentration provides a better chance for the insect to locate good food in the upper crown. We may suppose that weather and both forage and tree growth are related; but McLean and Smith (1973) were unable to find significant correlation between forage yields and tree ring growth. The high variation in yields and ring width behaviour is attributed to local conditions or microsite. We may assume that weather directly affects the tussock moth as well as tree growth. The insect also affects trees so that the impact on tree growth is likely a summation of at least climate and defoliation. Trees affect the insect in a feedback fashion: - 89 -As Koerber and Wickman (19.70) emphasized, an outbreak may only magnify the impact of the weather if the weather correlated with the outbreaks retards tree growth. The Duff and Nolan (1953) concept is not helpful in isolating impacts of several extrinsic.or "environmental";.^factors such as insects and weather. So I examined impacts on tree growth by analysing ring growth behaviour in Ponderosa pine and undefoliated Douglas-fir trees, as controls, and in trees defoliated to different degrees. Various workers have examined tree growth following silvicultural treatments such as pruning, thinning and fertilization, and following natural and man-made, disasters of defoliation. Effects of pruning, were examined, in Douglas-fir. by S;€ein v(,1955) -and Staebler (1963; 1964). Pruning from below up to 25 percent of the live crown increased diameter growth.at breast height, indicating the pruned branches were a burden to the tree. Increased severity of pruning almost invariably reduces growth. In Lodgepole pine defoliated by the Lodgepole.needleminer, Coleoteehrvites (Evagora) starki, reduction of radial increment is immediately evident in upper parts of the stem, but there is a two year, lag at breast height (Stark and Cook, 1957). For Grand fir, Douglas-fir and Engelmann Spruce defoliated by western Spruce budworm, Chor-Lsteneura oecidenetaliss C.B. Williams (1967) found parts of the stem near the ground were.more "complacent" in exhibiting reduction than parts higher up. Douglas-fir showed the least reduction. Although generally tree growth is directly proportional to amount of foliage present (Mitchell, 1975), it isr; necessary to evaluate, the feeding behaviour of a defoliator, and the needle crop distribution - 90 -within the crown configuration in order to postulate some theoretical expectations for impacts of defoliation. Radial growth impacts include missing and discontinuous rings (O'Neal, 1962; 1963), immediate and delayed reduction in ring width, no response and increased ring width (Staebler, 1963; Polge and Garros, 1971). Polge and Garros explained the increase as due to mobilization of stored food in parenchyma. But cambial activity (and hyperactivity) is due to growth regulators rather than stored carbohydrates (Kozlowski, 1969). Webb and Kilpatrick.(1976) found significant reduction in starch content in Douglas-fir trees defoliated by the tussock moth near my study area. K. Graham (1963) was of the opinion.that increased radial growth is likely due to lower rates of transpiration and translocation following defoliation. Except.in extreme cases, temperature in the stem is unlikely to increase fast enough.to negate the increase in growth by promoting faster, rates of metabolism. Since growth in Douglas-fir is determinate, and since in this species most photosynthates are stored in buds and needles, effects of defoliation in one year should not be evident in earlywood the following year. Effects.should, however, be evident in latewood because its formation largely depends on current year photosysthates. Tables 16 and 17 summarize data of tree radial growth at breast height in 1976 at Mountain View. Defoliation occurred there in 1975. I have attempted to use 1974 and 1975 radial.growth in defoliated Douglas-fir trees, and 1976 growth in undefoliated trees as controls in examining possible reduction, in radial growth at breast height following defoliation. Ponderosa pine gained more radial increment in 1976 than in - 91 -Table 16. Average percent change in tree radial growth at breast height following defoliation at Mountain View in 1975. Top: Growth in 1976 as percent change from 1975 growth in the same trees. Bottom: 1976 growth in defoliated trees as percent of 1976 growth in undefoliated Douglas-fir trees. Ponderosa pine Douglas-fir Deg. of defoliation: 0 0 Low Medium High Lw +50 -44 -19 -27 -39 Ew +17 -61 -26 + 8 -22 Ring +24 -57 -25 - 3 -27 Lw/Ring +21 +29 + 7 -25 -17 Lw 76 146 94 Ew 98 242 155 Ring 90 211 138 Lw: latewood; Ew: earlywood. Data are based on 16 cores in each class. - 92 -Table 17. Average percent change in tree radial growth at breast height by tree size, following defoliation in 1975 at Mountain View. Growth in 1976 as: Top: Percent change from 1975 growth in the same trees; Bottom: Percent change from 1974 growth in the same trees. Diameter class (cm) «15 . 15.1 - 25-.0 25.1 - 35.0 >35 Deg.iiofree • . iJdefol^vdei.Lw;' Ew R Lw Ew R Lw Ew R Lw Ew R P 0 +107 +70 +78 +19-8 -3 +5+7+7+83 +23. +35 y '  ' DF 0 - 60 -42 -50 -79 -83 -82 +21 -47 -32 - 51 -69 -65 L - 37 -55 -50 -32 - 9 -14 - 7 -38 -30 - 15 -17 -17 M -3-6-0.2+41 +70 +64 -54 -20 -29 - 38 +5 - 3 H - 40 -28 -32 +41 -0.03 -11 -38 -38 -38 - 32 -37 -36 P 0 +178 +79 +98 +29-6 +0.4+7 -20 -15 +111 +42 +56 y DF 0 -60 -51 -53 -67 -822 -79 - 2 -33 -24 - 36 -58 -53 L -20 -14 -.46 + 2 -16 -12 0 -27 -20 + 8 -20 -17 M + 8 -12 - 8 +62 +21 +28 -57 +11 -13 + 2 +30 +22 H -20 -18 -19 -8-0.2 - 1 -47 -31 -35 - 51 -44 -46 Lw: Latewood; Ew: earlywood; R: Total annual ring. Data arelbased on 4 cores in each class. - 93 -1975. The opposite is true for. sympatric Douglas-fir. In defoliated trees, within the first year, reduction in radial growth was less than in undefoliated Douglas-fir trees. The reduction is apparently not correlated with tree size. Defoliation seems to have retarded immediate growth reduction. Evidently at breast height, there is a lag of more than one year in Douglas-fir growth- following defoliation. Impacts at breast height may be more dramatic in characteristics other than radial growth. It may be speculated that defoliation early in the growing.season results in nutrients being rechannelled from branch tips to the trunk. In the past century., trees in the study area experienced two periods of good growth and three of poor growth (Figures 25 to 32, inclusive). These graphs and significant correlation coefficients of 0.6, 0.8, 0.7 for earlywood, latewood and ring width of Douglas-fir and Ponderosa pine indicate that the two species experienced similar radial growth trends. Generally, Douglas-fir grew faster than Ponderosa pine trees. But during periods of depression, growth of Douglas-fir was reduced more than Ponderosa pine. Superimposing North Thompson.and Kamloops historical out breaks (Figure 1) on the radial growth behaviour graphs, it is note worthy that three outbreaks occurred during low growth periods (depres sions), and one during a isemifavo'fable period;!. A depression may accommodate more than one outbreak. Because we know nothing about the specific.outbreak history of these trees, we cannot state whether or not defoliation played a role in the growth depressions. It is not even known if the outbreaks occurred.in the general vicinity. Some . A , - 94 -growth depressions precede the beginning of.outbreaks, indicating that growth reduction in Douglas-fir is not necessarily a result of outbreaks. Probably some common factor(s) affects radial growth behaviour in the two species. The factor is probably a macro one since trees used in this historical analysis are from a wide area. The coincidence of some outbreaks with depressions in tree growth suggests a similar factor also affects insect populations. But apparent inconsistency in the coincidence suggests a macro factor is not the only important one. The overall factor may.set the stage within which several cofactors function to determine insect population trends. More conclusive statements cannot be made from this analysis without the risk of making tenuous assumptions. A knowledge of history of the trees or stands studied here would have made this part of the project more than exploratory. The difficulties of assigning cause-effect relationships when enough information is not known are evident in the following chart. Probably the most spectacular impact, of Douglas-fir tussock moth outbreaks is the sometimes extensive, visually obvious tree mortality they inflict on.the landscape. In Forest Insect Disease Survey records, reports of extensive mortality near Vernon were recorded in the following years: 1921-1922, 1929—1930, 1938-1939, and 1945. Mortality near Chase, Armstrong, Hedley and other places was noted in various years. Many workers in the U.S. also have expressed.concern about the destructive power of the Douglas-fir tussock moth. D.A. Graham (1974) stated that the insect can cause mortality one year following defoliation in unnamed host, and contrasted The two species respond similarly to the same or different factors Yes Working hypothesis: Radial growth behaviour and outbreaks likely respond to a common ecological factor(s) Other factors play important roles in outbreak occurrence Working hypothesis: Radial growth behaviour and insect populations are determined by, or respond differently to the same or different ecological factors • No ± Lack of correlation may be due to high variability in the population: Regroup trees into smaller populations or increase sampling intensity Trees are complecent: Possibly the hypothesized inter action between climate, trees and insects may be valid Impacts may be in terms of other than radial growth behaviour. - 96 -this with Balsam fir which can tolerate as many.as three consecutive years of defoliation by spruce budworm. In 1949, a Forest Insect and Disease Survey report noted that in Monte Creek and Duck Ranger Districts, defoliated trees between 35.5 and 40.6 cm dbh were also.attacked and killed by the Douglas-fir beetle, Dendrootonus pseudotsugae. In western U.S.A., a range of mortality from 66 percent to 95 percent has been attributed to bark beetles in tussock moth defoliated:stands (Wlckman, 1958; 1963; Wickman et^ al., 1973; D.A. Graham, 1974; Carolin and Coulter, 1975). Size "preference'', by the beetles was not indicated in the FIDS reports; but it was evident in my study area (Figure 37). The 1949 Forest Insect and Disease report also suggested that Douglas-fir is highly resilient to tussock moth defoliation. It stated that all (Ponderosa) pines which had been defoliated, presumably by the tussock moth, were dead in contrast to Douglas-fir. which recovered quickly. A 1964 Forest Insect and Disease report: "... (Douglas-fir) trees which were almost denuded (of foliage) in 1963 grew a surprising, amount of foliage." More recently, Carolin and. Coulter (1975) have shown that in the U.S. Pacific Northwest, Grand fir is damaged more than Douglas-fir even when insect density is. equal in the upper 1/3 of the crown in both species. The authors did.not speculate on possible, reasons for this difference. Is it due to certain differences in lengths.of time needle crops are retained on the two species, or to differences in foliage (age) distribution in.their crowns? Wright (1974) had observed that actually the greater damage to Grand fir stands was due to differential defoliation by the tussock moth. But to state, as he did, without Figure 37. Some defoliated Dougi" - 97 -Figure 37. Some defoliated Douglas-fir trees infested with bark beetles. Size preference evident. Note thriving Ponderosa pine. further comment that susceptibility increased with Douglas-fir component in the Blue Mountains seems contradictory, and serves to confuse the situation. The question of what level of acute or chronic defoliation is necessary before Douglas-fir trees succumb is relevant in assessing mortality impacts more precisely. In the Pacific Northwest, fifty percent defoliation was indicated to be the threshold in unnamed species To face page 98 Figure 38. Lush current year needle crop on small Douglas-fir trees, following complete defoliation during the previous year. Evidence of good recovery unless redefoliated! Top', middle: Cherry Creek, Bottom: Indian Gardens - 99 -U.S.D.A., 1973b). I did not.embark on an estimation of a threshold because of difficulties involved in estimating defoliation levels precisely. Instead, I examined recovery of Douglas-fir trees which had been completely defoliated in the preceding year. Color and position of current year needles facilitated my classifying a tree in this category. Good recovery rates are evident in small trees (Figure 38) , and mortality is most serious only among large trees (Tables 18, 19, 20). Table 18 Recovery of completely defoliated trees in a stand at Heffley Creek hillside (M-site quality), and dynamics of their growth. Data are based on. 170 trees from 2 representative strips. Trees were tallied one year following defoliation. Diameter class (cm) <5.0 5.1-10.0 10.1-15.0 15.1-20.0 20.1-25.0 % Survival 76 72 80 66 50 Avge diam (cm) 2.8 7.1 11.4 17.5 23.1 Avge height (m) 2.6 7.0 10.6 13.4 16.5 Avge Cr. length (m) 1.5 3.8 4.4 5.4 6.0 Avge Cr. width (m) 1.3 2.3 3.4 4.6 5.5 Diam/Cr. lgth 1.9 2.0 2.6 3.2 3.9 Diam/Cr. width 2.2 3.1 3.4 3.8 4.2 Ht/Cr. lgth 1.7 1.8 2.4 2.5 2.8 Ht/Cr. wdth 2.0 3.0 3.1 2.9 3.0 - 100 -Table 19 Recovery of completely defoliated trees on a southeast facing dry site, and dynamics of their growth. Data are based on 158 trees - residuals from salvage logging. Trees were tallied one year following defoliation. Diameter class (cm) <5.-0 5.1-10.0 10.1-15.0 15.1-20.0 20.1-25.0 % Survival 76 68 76 82 66 Avge diam (cm) 3.1 7.6 12.5 17.3 20.8 Avge height (m) 3.0 8.3 11.7 13.4 16.7 Avge Cr. length (m) 1.6 3.3 6.1 8.2 8.3 Avge Cr. wdth (m) 1.3 1.7 2.3 2.7 3.4 Diam/Cr. lgth 1.9 2.0 2.1 2.1 2.5 Diam/Cr. wdth 2.4 4.5 5.4 6.4 6.1 Ht/Cr. lgth 1.9 2.2 1.9 1.6 2.0 Ht/Cr. wdth 2.3 4.9 5.1 5.0 4.9 Table 20 Survival of all, completely defoliated Douglas-fir trees tallied in the study. Data are based on 466 trees ffrom most plots and.strips in the study area. Trees were tallied one year following defoliation. Diameter class (cm) • *.fl* ' 5 i-.T-rr.n- lffj-i'5 ij 15 i-?r ' , . <5 .0 5.1-1070 1011-T5.0 -15 .1-20.0 21."l-25.0 >25.1 Trees tallied 152 136 109 44 17 8 Trees living 113 98 79 37 12 4 % Survival 74 71 64 84 71 50 - 101 -Apparently, the larger the tree.the lower the chances of recovery as levels of defoliation increases. Older stands traditionally have low stocking.levels; increased mortality following defoliation further reduces stocking to lower levels. This is advantageous from the point of view of graziers as increased understory forage yields may be realized. The ratio-of dbh/crown length or width increases with tree size, indicating the crown becomes relatively smaller. The ratio is higher on drier and poorer sites, where a smaller crown may serve to minimize evapotranspiration. In many stands in the ecotone, trees are widely spaced so that competition seems insignificant. The chance of some trees being released following defoliation of others is remote, except in a few thick patches. Vincent. (1962) observed release of Balsam firs in stands infested with Spruce budworm in the well known Green River Watershed in New Brunswick. He thought the evidently slow release of advance Balsam, fir regeneration was due to more light penetrating through, and reduction of competition following defoliation. Advance Douglas-fir regeneration is not immune to defoliation by the tussock moth, but most of it recovers well. Harwood (1975) was of the opinion that severe Douglas-fir defoliation affects cone bearing trees and seed production for some time. Vincent (1962) also alluded to the need for seed availability immediately prior to defoliation for successful natural regeneration in Black spruce stands. In the study area, under dense stands and high crown cover tree regeneration was uncommon. The denseness of - 102 -these patches made them inaccessible to ungulates. The forest floor and litter layer remained undisturbed and thick. Reduction of stocking and density by defoliation would make stands accessible to ungulates. Scarification effects should be evident in increased forage yields and regeneration. In protected openings, high stocking of Douglas-fir regeneration was common. Ponderosa pine regeneration was very scarce even when cones were clearly abundant on the ground (ITable" 21, Figure 39). Whether.the Ponderosa pine seeds were eaten by mice and squirrels Table 21 Average regeneration stocking in experimental plots under canopy and in adjacent protected openings. Averages are based on 4 plots under canopy, and between 4 and 10 plots in.openings. Only the 8 tree plots out of 36 had regeneration. 2 Regeneration stocking (No./m ) Tree plot Adjacent protected opening Basal area %CC F Py F Py z o. moose (mz/ha) 13.8 39 2.0 - 0.0 16.0 64 1.5 - 2.5 16.0 63 0.5 - 7.6 -18.4 69 0.5 - 1.3 20.7 42 5.0 - 7.4 2.0 23.0 36 3.3 - 5.1 25.3 64 0.5 - 7.8 41.3 51 2.5 - 3.0 - 103 -Figure 39. Abundance of cones on the ground; but no regeneration was evident in some openings. Behind the range pole is undefoliated plot Forage yields (kg/ha): Yfc = 51.6; %CC = 67 67 , 64, . Cherry Creek, 1974. m lens ' more than were Douglas-fir seeds, or whether there were differences in seed viability between the two species is not known. The low regeneration stocking under high crown cover regardless of the stand density implies that light is limiting for Douglas-fir regeneration in the ecotone. For adequate natural regeneration, seed availability prior to defoliation appears necessary. - 104 -Salvaging dead or dying trees reduces loss of timber volume and value. Unlike the process of salvaging bark beetle infested trees salvaging tussock moth killed trees does not drain insect numbers from infested stands. It may in fact serve as a means of dispersal for the insect if salvaging is carried out during the summer month's".-The ability to salvage is constrained by size and pattern of an outbreak, which in turn affect economics of harvesting. Except in second growth, accessibility is a costly problem especially for the small (gypo) operator. The age and size of infested stands, and of dead trees also determine the feasibility of salvaging. In California, extensive outbreaks in true fir stands often result in a race against Scolytus and Tetropivm beetles, and decay. Hidden defects of wetwood (Wickman and Scharpf, 1972) are also important. In a California study, 81 percent of topkilled trees had wetwood along the whole tree compared to only 2 percent in uninfested stands. Wetwood reduces timber quality, as it leads to excessive checking and collapse, and gluing difficulties due to uneven distribution of moisture at equili brium following drying. The race against time also is serious when logging is restricted to certain times of the year, due to excessive snow, wet ground or other reasons. In my study area, if salvaging were restricted to dead trees or patches harvesting costs would probably be prohibitive unless those people who benefit from increased forage yield contributed. Quality of dead Douglas-fir trees is probably not impaired within the first year following defoliation. Most of the logging in devastated stands is selective inasmuch as only trees larger than economic marginal size are removed (Table 22; -. 105, -Figure 40). The stand structure is consequently changed. Many large undefoliated Ponderosa pine trees are removed together with Douglas-fir trees for economic reasons. As much as 16 percent of the volume removed from several settings by an operator was "mix". Utilization stands were quite low. A cruise in stands adjacent 2 to a setting indicated a gross volume of 178m /ha. From a volume/ weight ratio, Balco Industries, a local firm estimated as having 3 extracted only 31m /ha, or about 20 percent of the volume. Such yields are due to low-poor site quality, possibly stand immaturity and the fact that only salvageable material was removed. Experienced cattlemen were of the stern opinion that skidding or tractor logging serves to scarify the soil, and is beneficial because this is likely to increase forage yields even without range seeding. Table 22 Stand composition of Douglas-fir trees in defoliated stands before and following salvage logging. Data are from 4 representative strips near Heffley Creek. Diameter class (cm) <5.0 5.1-10.0 10.1-15.0 15.1-20.0 20.1-25.0 >25 Percentage of trees in class Before logging 6 25. 25 23 2 19 Following logging 26 34 31 7 2 0 To face page 106 Figure 40. Structure of residual stands following salvage logging. Marginal tree size Ca. 20 cm. Top: Heffley Creek, public property. Bottom: Mountain View, private property (Mr. Inskip). Logging progressing into the devastated area. - 106 -- 107 -Salvaging may have impacts on the bird community. Birds of the primitive coniferous forest are almost all insectivorous. Removal of the forest enhances establishment of mountain and western bluebirds and other graminivorous birds, many of which prefer inhabiting coniferous forest - grassland ecotones (Thomas and McCluskey, 1974). They are mostly cavity nesters, and heavy herbaceous flora provides desirable cover. To the extent that snags may be critical for survival of these birds, care should be taken during salvaging to preserve some dead trees. Some insects may have a serious negative impact on bird populations as much as birds have on the insects. Defoliation may expose bird nests to the extent where ^nsolationmmay'be lethal-to ' nestlings, and may lead parent birds to abandon their pregeny. This impact was mentioned in a Gypsy moth, %yndnhc.ia (Porthetria) dispav, outbreak in eastern U.S.A. (Commonwealth of Massachussetts, 1908). K. Graham (1963, p. 251) refers to some European literature on this impact. Epidemiology Regulation of insect numbers reflects, in part, an impact of the insects on themselves. Knowledge of insect epidemiology enables economic entomologists to forecast trends of insect populations, and future damage so that appropriate action may be taken. The knowledge is necessary for formulating rational pest control policies and strategies. - 108 -Life cycle and sexual.dimorphism of the Douglas-fir tussock moth have already been mentioned in a previous section. Several studies have been undertaken in examining aspects of population dynamics of this insect. The cyclical nature of outbreaks in several locations is quite striking. So is the synchronous nature of outbreaks between locations over a wide latitudinal range from New Mexico to Kamloops. This distribution in time and space, and the wingless nature 4 of (j) adults make it unlikely for these populations to be related genetically. For populations which are close to each other, inability of theO^ to fly reduces frequency of gene flow between them. Genetic relationship would require consistent dispers al of c?c?, which is highly unlikely for populations farther apart than 2500 km - between New Mexico and Kamloops! Evidence indicates most populations are "independent" of each other. Even within a stand, growth of one population is not necessarily related to trends in another population. Mason (1974) has shown this in fifty plots in a 121ha stand. He employed R.F. Morris's (1963) analysis.of Trend Index [I = ^ /N ^] and concluded that for populations within a stand, the first, second and third year, the trend indices were 7, 3, <1 respectively. He also showed in a dynamic population model, N = f (density), that more than 80 percent of the changes in numbers during one year were due to changes.in the same population, during, the preceding year. This meant that contagion was insignificant or. immigration equaled emigration in numbers and quality. 4 Dr. R.R. Mason, For. Sc. Lab., Corvallis, Oregon, U.S.A. Concurs with this view in personal correspondence. - 109 -In spite of the fact that dependence has not been demonstrated, views implying it ' exists are not uncommon in the popular literature. Opinions such as "... the tussock moth can just eat its way to those firs on our coast,and "... District Forest Ranger said the (Douglas-fir tussock) moths were travelling toward a large stand of virgin timber..." are ill founded. They represent a misunderstanding of an important issue. We know the tussock moth does not swarm like locusts! The historical frequency of the tussock moth on the coast is negligible. In fact the insect may have never inhabited much of the coast. Unfortunately the alarmist views sometimes serve the: purpose of marshaling public support and sympathy for what would otherwise be unjustifiable control measures. Livingstone and-Tunnock (1973) were convinced . that surveys clearly showed great potential for the insect to spread. It is not clear whether they were worried about devastation of susceptible stands by autochthonus or "migrating" insects. Despite t^is.'amb'iguitNy,, the belief was used in an argument in attempts to persuade the U.S. Environmental Protection Agency to grant permission for the latest D.D .T. - aerial. spray in the Pacific Northwest (see U.S.D.A. 1973b). In Western U.S.A., where the tussock moth has been intensively studied, three population phases, viz. release—>peak—>decline are so commonly referred to as to suggest they are a constant feature of the insect. The idea and terminology appear to. have originated from ^ The Oregonian. 3rd November, 1972. The Oregon Journal. 18th July, 1947. (Courtesy of Dr. R.R. Mason, personal correspondence). - 110 -Greenbank (1963) in his work, with eastern Spruce budworm. Most tussock moth infestations are recorded as lasting three years (U.S.D.A., 1973a) - hence the so called three year cycle. Wickman ejt al_. (1973) ascertained the cycle in five separate case studies. Each phase is considered to last one year. This time dimension may serve to warn resource managers about the state of an outbreak, but it is not likely representative of all populations. The preparatory (release) phase probably lasts longer than one year. It is difficult to define what constitutes a release phase because the very low population levels imply difficulties of detecting the insect. A "shotgun effect" (Figure 41) so common on infested landscape suggests the three year cycle may be unrealistic. It is evident in Figure 1 that outbreaks may last for as long as five years. As D.A. Graham (1974) pointed out, in one.infested patch the cycle may be a three year one, but it may be longer in a stand or forest with several patches. The infesta tion patch we see up the hill is not necessarily in the same phase as the one down the valley. The quasisynchronous occurrence of outbreaks over great distances leads one to suggest that some regional factor determines, at least in part, the outbreaks. If this were true one would suspect effects of the same factor, to be expressed in other organisms. If this subsidiary postulate were shown to be true, it would lend credence to the principal hypothesis. Rejection is not "derogatory" to the primary hypothesis, however (Peddie, 1938). This area is examined through tree ring analyses reported elsewhere in this thesis. The Douglas-fir tussock moth is a "hitchhicker" (Wolfenbarger, - Ill -Figure 41. Shotgun effect: typical nature of Douglas-fir tussock moth infestation on a landscape. - 112 -1946) on people, animals, automobiles and in wind, in the larval stage. Probably mortality is high in the transportation medium unless travel is restricted to the terrestrial zone (Berland, 1935), or active plankton zone (Wellington, 1945) - more than 2 km above ground and 7°C. The reciprocal influence of trees on the tussock moth indicated in the interaction triangle on page .88 is real. It represents "resistance" or feedback common between living systems. I have encountered in the literature two sets of data which although not intended to illustrate this concept, illustrate it well in the tussock moth situation. Condrashoff and Grant (1962) reported their survey on distribution of diapause sites (larval coccoons) in stands defoliated by the tussock moth near Vernon, B.C. In a stand with trees denuded of their foliage, the understory vegetation is favored for oviposition sites; Douglas-fir regeneration being the preferred plants. Number of cocoons on the reproduction averaged more than twice the number on Choke cherry and Douglas maple combined. In a stand where the overstory is completely defoliated, emerging larvae immediately feed on - the reproduction. The distribution of cocoons within a Douglas-fir tree varies with the degree to which the tree was defoliated (Table 23). In general, Luck andDBahlsten'(1967) arid Dahlsten et^ al. (1970) concluded similarly for cocoon distribution in White fir. Heavier defoliation results in a shift in relative "preference" of the insects for the bark and lower parts of the tree as oviposition and refuge sites. On these sites and on the favored understory vegetation, probably predation and other mortality factors are more intense. The distribution also illustrates the need for - 113 -Table 23 Distribution of Douglas-fir tussock moth cocoons in lightly and heavily defoliated trees. (Modified from Condrashoff and Grant, 1962). Crown stratum Upper Middle Lower Bark Degree of defoliation Light Heavy 189 156 . 74 196 80 223 16 139 stratifying the sampling design along the vertical axis for purposes of sampling tussock moth populations. Also it is necessary to change the intensity of sampling in various parts of the crown and tree as the degree of defoliation changes. By enhancing air circulation and formation of thermals in the forest, intensive defoliation by the tussock moth promotes its chances for dispersal. This improves the chances of survival for both emigrants and residuals when food is still available. But this may also reduce the residual populations to a point where biotic agents can bring them down, especially when stress is prevalent in the residuals. Current year needles are eaten first during the release phase. During peak and decline phases only older needles which probably have lower food quality are available. Beckwith (1976) free-fed and force-fed - 114 -tussock moth larvae on foliage from.different crown strata of three hosts. I tabulate his data to illustrate a reciprocal impact: Host Source of foliage % Survival *Free feeding *Forced feeding Douglas-fir Top 100 100 Bottom 100 100 Grand fir Top 100 100 Bottom 100 90 Subalpine fir Top 90 80 Bottom 100 70 *Free feeding: On current year foliage *Forced feeding: On older foliage. Forced feeding simulated the situation following release phase when mostly low quality needles would be available. The low nutrient quality and probably relative quantity in the older foliage was indicated by high frass production. Exhaustion of food may be suicidal when dispersal is not possible. Exhaustion of high quality food reduces survival rates, (see also Mason and. Thompson, 1971), and probably quality which may be reflected in fecundity. Food quantity affects size of adults, thereby their fecundity. Leonard (1970) advanced a provocative hypothesis on several aspects of population regulation in the liparid, Lymantria dispar. He reasonably.suggested that starvation enhances dispersal. But to state - 115 -that "at high density dispersal., is induced prior to the crush of the population" is misleading because it implies dispersal is "induced" socially.or otherwise from within the population. Yet it is clear that environmental forces are necessary for dispersal. He also suggested that during evolution, loss of wings in the Oj^ was accompanied by population flushes or wide fluctuations. The flushes appear self destructive, but because starvation enhances dispersal, they are a mechanism which ensures gene.flow.and maintenance of variability in the populations. At high density dispersal is induced before the population.crashes and this sows the seeds for new populations. This erroneously implies that whenever we detect a population crash, the insects have actually gone elsewhere. Whether or not loss of wings in Douglas-fir tussock mothQ^ was compensated for by development of dispersal mechanisms in larvae, the loss has not been very costly as gene flow and mating are facilitated by the being almost stationary and producing a sex pheromone, and the hairy nature of the larvae for aerial dispersal. Several biotic agents have been recorded with increasing frequency in the field during the decline phase. Most of them have been identified from rearing projects. Apparently, the most important agent is a.nuclear polyhearal virus (NPV). In Forest Insect and Disease Survey Annual Reports, a condition.was described in 1939 as a wilt disease; in.1945 as.a virus; in 1954 as a polyhedral virus, and in 1955 as NPV. In 1952, a biological control attempt with a virus -probably NPV - was undertaken against the tussock moth in B.C. NPV has also"been implicated in decline of several outbreaks in the U.S.A. - 116 -Through a life table - factor .analysis approach, Mason and Thompson (1971) found for a virus d^F.of 41 percent. Hughes and Addison (1970) have identified two strains of NPV in the tussock moth. The virus prevents oviposition by <j) tussock moths (Dahlsten, et al. , 1970): it is not clear whether this is due to effects on oocyte development, hormonal control of sexual development and activity, or behaviour. Martignoni e_t al_. (1969) found a cytoplasmic polyhedral virus in Douglas fir tussock moths. Where NPV has been claimed to cause significant mortality, it has often done so after severe damage has already been inflicted by the tussock moth. The virus appears to let the insect get out of hand. Long lags are probably characteristic between populations of these two organisms. Stress may be necessary before the pest succumbs to the virus. Is stress provided by food shortages, and increase in pest density? Is high density required as a sufficient means of disease transmission in the population? Notwithstanding our lack of understanding of such points, NPV is registered now in the U.S.A. for use against the Douglas-fir tussock moth. Circumstancial evidence indicates a red ant, Formica intergroides, protects trees from defoliation by the tussock moth. Active ant colonies (Figure 42) were encountered in the study area. Their scarcity and uneven distribution in affected stands were apparent. Macroclimatic factors related to aspect and elevation, and microsite factors related to stand structure may limit establishment of colonies in the study area. Hughes (1975) showed such environmental factors to be limiting for colonies of two insect predators, Formica rufa and To face page 117 Figure 42. Active ant colonies. Top: Cherry Creek - 1975; colony behind the range pole. Bottom: Mountain View; colony behind the log piece. Note some defoliation in adjacent trees. - 117 -- 118 -F. lugubris in Wales. Formica intergroides appears to "displace forcefully tussock moth larvae from branches.^ In the Coniferous Biome of North America some of the largest groups of birds feed on leaf eating insects. Orioles, Vireos are common in the interior of British Columbia, but their predatory influence may be insignificant in reducing insect numbers. Turc'ek (1948) in Checkslovakia noted that toxic gypsy moth larvae are eaten by birds including orioles and starlings. In-Germany, hole nesting and other birds also eat gypsy moths (Luhl and Watzek, 1976). So the presence of a toxin in Douglas-fir tussock moth larvae may make them poor, but not unlikely candidate for birds' food. Cost of an outbreak A decision to protect or to not protect forest resources has primary and external benefits as well as costs. Externalities are an important characteristic of the forest resource and decisions - governing its management. Traditionally, managers measure quantitatively these benefits and costs, commonly in monetary terms. In a managed forest', direct costs of an outbreak are real as they represent assaults on capital investments made in it. In a natural forest, on the other hand, losses are real only to the extent that we recognize the current market value of the crop, and the cost of time during which the crop occupies the land. Time is not a free resource: this concept is acknowledged in all investment decisions in part by the use of discount rates. This ^ Observation by Dr. R.F. Shepherd, personal communication. - 119 -is true also for conventional cash flow stand analyses. Clearly, value losses are sustained whether damage is done to natural or man-made stands. Investments to reduce these losses are essentially not different from any other in forest management. Because forest protection must compete with other cultural treatments for scarce funds, there is need to express costs and benefits in terms of a common parameter. This is necessary also for analysing feasibility of, and ranking investment projects to decide between alternatives. How do we actually measure losses and benefits in an outbreak? We face two basic problems here: (i) identifying sources of benefits and costs, especially external ones, and (ii) quantifying them. The sources may be more complex than was indicated by Stark (1975). Difficulties involved in quantifying benefits and costs, especially external ones such as soil protection and aesthetics, were evident at the 1974, 25th Annual Western Forest Insect Work Conference, at Salt Lake City, Utah, in discussions by several resource economists (Curtis, 1974; Michalson, 1974; Rivas, 1974; R.G. Williams, 1974). These difficulties are real; yet there is undeniable need for undertaking economic analyses to ensure maximum returns from investments. Economic analyses of specific insect problems are scarce in the literature. As we are unable to quantify convincingly many costs and benefits, the best we can do is use classical criteria of Benefit-Cost ratio, Internal Rate of Return, Net Present Value only as guides in the investment decision. Many forest protection decisions are made on a "gut feeling", or for political, strategic or speculated reason such as the argument that fire hazard is a problem in defoliated stands. - 120 -In B.C., the public, as the major owner of the forest resource, bears most of these losses. But some segments of society sustain higher immediate losses than others. The cost of an outbreak to a forest worker represents reduction in income as a result of depletion of the resource following defoliation. The cost of an outbreak to different workers may vary because of in situ differences in their income. The more income one foregoes the higher the cost. For an individual who becomes ill from conjuctivitus, dermatitis, pulmonary and other ailments caused by toxins from the base of tussock mothl'.s urticating hairs (Gilmer, 1923; U.S.D.A., 1973b; personal observation ), his costs may be significantly higher. The cost to a grazier may even be offset by increased forage following defoliation. Consider loss of a Christmas tree crop at Pritchard, B.C. (FIDS, 1948), and of shelterbelts around farms and residences in Idaho and Washington (Tunnock, 1973): the tree farmers and residents did not sustain equal losses. For a forestry enterprise which is dependent on backward integration of the resource for survival and competitive ability, an outbreak is a matter of life or death. These differences in sustained costs stem mainly from the externality characteristic of the resource, and its specific ownership pattern. In landscape architecture and urban forestry, quantifying value of a damaged tree is equally controversial. The value may be equated with the amount of money an owner is prepared to pay in order to save the tree. But various owners would spend different amounts of g I witnessed a faller complain of serious skin irritation while logging a devastated stand owned by a Mr. Inskip, at Mountain View. This was in the stand shown in figure 40 - bottom. - 121 -money for the same tree. It is generally agreed that one of the follow ing is the real loss to insect kill: (i) the current market value of the tree regardless of historical costs; or (ii) the cost of establish ing a new tree up to the age, size and value of the dead one immediately prior to its destruction. These concepts are often used in forest property compensation. Aesthetic values of trees in the wilderness are even more difficult to quantify. The value of trees in a park may be assessed by visitor use. But what is the value of trees along a public highway? Evaluation of values here becomes a problem of evaluating mainly social attitudes. Wickman and Renton (1975) explored two simple methods of evaluating cost of one outbreak in a Pine-Fir campground with 8 units, in California. (i) Following tree mortality, cleaning up was required to reduce fire hazard, and danger to camp users. Felling of dead and dying trees, and slash disposal cost $90.00. Topping top killed trees cost $100.00. The total cost was $190.00 ($23.75 per unit), (ii) There were 370 trees in the campground, or 46.2 trees per unit. The replacement value of each unit was given as $1500.00. Assuming the Pine and Fir trees have equal aesthetic appeal, the value of each tree equals $33.00. A total of 25 trees were killed by the tussock moth outright: total value lost in mortality equals $825.00. To this should be added the cost of tree felling and topping ($190.00) • from (i). Total loss amounts to $1015.00 ($126.00 per unit). - 122 -The first procedure is unacceptable as it ignores replacement costs. Both procedures unrealistically ignore possible reduction in visitor use related to lower aesthetic value, and annoyance caused by the tussock moth. I submit that the real cost of an outbreak is heavily dependent on "whose horse is getting gored". Recently a new concept called Allowable Cut Effect, dubbed ACE, has appeared in the forestry literature (Schweitzer ejt al. , 1972). The concept is not subscribed to unanimously. The concept: in a regulated forest with some old growth where yields are constrained artificially and regulated, any silvicultural treatment which increases yield in second growth should justify our reaping benefits of increased yield immediately in the old growth. The immediacy of these benefits results in highly favorable Internal Rates of Return. ACE is real, and is demonstrable in conventional stand analyses. The fact that it can be traced to in situ artificial constraint on the yield does not nullify its reality. Strictly on the basis of conventional investment cash flow analysis applied to protection of forest inventory, Bell e_t al. (1975) implied that we often exaggerate the cost of an outbreak. While ACE is used to promote some stand improvement investments, it can be used to discourage others such as protection of inventory: "...like the two edged sword, ACE can cut in both directions." Bell et al. (1975) used data from a case study of the latest tussock moth outbreak in the Umatilla National Forest. The U.S. Forest Service had estimated the damage there to be over two million dollars. But Bell and his associates argued as follows:- In that outbreak, nonsalvageable mortality - 123 -amounted to about 1.2 million board feet (fbm) every year for the next rotation of 115 years, or 1.2 million fbm -''"of the allowable annual cut in the management unit. The loss was calculated according to the familiar Hanzlik formula, which is now obsolete in that area. At the market value of $62.00 per thousand fbm, the loss was $74,400.00 annually. In cash flow investment theory, this represents a series of annual payments for a terminable period of 115 years. Assuming a reasonable discount rate of 10 percent, and using the appropriate 9 investment model to discount to the present, the loss amounted to only $744,000 (cf. more than $2,000,000.00). only by the present generation. It is equivalent to using zero discount rate. As in depreciation theory, these costs should be spread over a meaningful period - a crop rotation in this case. Insect damage may not result in as much losses as we often think. Because of higher timber values and shorter rotations on better sites, discounted losses are likely to be higher there. Such sites deserve higher priority in protection. Losses on poorer sites may be very low; benefits accruing from increased forage production following defoliation make the losses even smaller. This is an argument for leaving outbreaks on poorer sites to run their course. If we invest money on such sites, the opportunity cost we incur may be too high. The younger the stand ^ V = r/i I(l+i)n-l[ where V = discounted value; r = annual payment; O I - . . \ n I Bell e_t al. did not give their model. Results obtained from this one were slightly different f rom theirs. Reduction of the discounted loss is indisputable! The undiscounted loss is so high because the cost is borne i = discount rate; 10% = 0.1; n = number of years-rotation. - 124 -when it is damaged, the less value we lose through mortality. For various reasons, discount rates lower than 10 percent are more realistic. But to exclude discounting at all is unacceptable. When outbreaks are recurrent, as they are within some management units, loses may overlap and magnify over a single rotation. Because values of stands differ by location and other characteristics, and because discount rates also differ in time, we obviously need detailed evaluation of benefits and costs for each specific problem.in-order ,to determine its real magnitude before attempting serious control measures. A blanket control project in the study area is unjustified. It may also be undesirable when we examine benefits accruing to forage production if an outbreak is left alone on some sites. u... involvement in resource protection from insects and diseases has been oriented toward protecting commercial timber values. Where these values are low ... but where watershed, wildlife, esthetics and recreational values of a :vigorous forest are high, we need to rethink our priorities." John R. McGuire, Chief, U.S. Forest Service, 1976. Resource use conflicts and pest management strategy It is unwise to advocate pest management strategies without considering the land or resource use pattern in an outbreak area. Mere visual impacts on trees should not necessarily lead to invoking of pest suppressive measures. Sometimes for various reasons it is more sensible to leave an outbreak alone unless it threatens.valuable stands. Where several resources are involved on the same land, as in the B.C. interior, it is important that we consider all resources, not only one. - 125 -It may be easier to decide on the strategy when the sympatric resources are compatible. In this section I will consider two apparently "conflicting" resources in the study area - forage and tree production -in developing a strategy for managing the Douglas-fir tussock moth. In my arguments, I will rely heavily on the data and foundation established in preceding sections. Inasmuch as land tenure is indicative of land use priorities, when protection of forest inventory is considered it is imperative that we consider the resource values and their relative importance in the whole. The relevance of forest land tenure - the interphase between forest law and economics - in forest resource use is apparent, especially in expanding scope of forest resource management. On the Canadian scene, most forest land tenures such as those in the public domain are indicative of relative magnitudes of resource values on the respective land. Within the Interior Douglas-fir Zone about 3.2 million forested hectares were being used for grazing twenty years ago. This included all the lower zonal ecotones. In contrast, there were 1.2 million hectares of open range most of them below the timber line (Sloan, 1956; Tisdale and McLean, 1957). More recently Pearse (1976) put the total area of usable forested range land at about 6.7 million hectares, most of which are in the Kamloops and Cariboo Forest Districts. Evidently, the interior forests are important sources of forage. In the open range, forage production is the dominant, and often the only tenured use. Tenure on forest land is variable. The study area is surrounded by several Public Sustained Yield Units (PSYU) and Tree Farm - 126 -Licences (TFL). All of my study plots were within the Kamloops PSYU (No. 31). Within the unit, some land is privately owned. P.S.Y.U.'s and T.F.L.'s are dominant at higher elevation, where wood and fibre production are primary uses, and forage rights are usually for one year in the form of grazing permits. In the ecotone, many stands are managed under 21 year grazing leases; here forage production is.dominant over timber production. But logging may be undertaken in emergency to salvage timber values through Timber Sale contracts while the lease is still valid. The forested land covered by these leases is administered under the Land Act, and is of low site quality for timber production. Many of the defoliated stands were on such land (Table 2). The low site quality is evident also in low timber yields extracted during logging. On most of the privately owned forest land, grazing is the dominant, but not necessarily the only use. Tnasmuchcasithe-public makes^various.ioverlapping demands on the ecosystem, conflicts in resource use and management are Abound to arise. One of the most serious and oldest conflicts here involves range and timber values. Professional resource managers, graziers and others have been involved in the conflict. As early as 1920 the pro vincial minister responsible for forest resources, the Hon. T.D. Pattullo, stated that of the forested land 3.8 hectares were potentially suitable for agriculture and would be put to farming (probably forage production) after timber harvesting. The withdrawal seems not to have materialized to a significant extent as there were 3.5 million hectares of forested range land thirty-six years later (Sloan, 1956). In his Royal Commission reports, Chief Justice Sloan (1945, 1956) considered the - 127 -issue of multiple use, especially in the interior, as one of the most (C controversial in management of the provincial forest resource. It was during the 1955-1957 commission hearings that the controversy between range and timber interests reached its apogee. "It is chiefly in relation to grazing on crown forest land that conflicts of interest arise. (The use of) the land for production of timber, cattle, the extraction of minerals and available water ... will create ... unfortunate, although presently avoidable consequences". (Sloan, 1956; p. 681). It is fallacious to believe that resource conflicts are due solely to mismanagement, contrary to the Argument of Crown Counsel, section 199 (Anon. 1956): "One witness said /%,±tia%ty'' that there was room for both loggers and cattlement if the matter is properly managed." In a brief, a Mr. T.G. Willis, an agrologist with the then Dominion Agriculture: "Their (graziers') other problem is the encroachment of the forest on their grazing land. Stockmen have been agitating this matter for a number of years, so far without any tangible results." "Jack (Lodgepole) pine, willow and alder" are blamed for the forest encroachment. In concurring with these opinions, Sloan called these trees trespassers, and erroneously considered the encroachment problem more serious than overgrazing. The same attitude had been set in his earlier report - (Sloan, 1945; p. 164). But as forester Alan Moss testified: "We (the interior forest industry) are just as touchy about encroachment of (sic) forest land by grazing land as the cattle industry is on the encroachment of grazing land by forestry" (Sloan, 1956; p. 699). - 128 -In view of the broad frame of reference for the commission, it is surprising that equally important questions of forage yields and quality, timber yields and soil conservation influences by ecological forces (e.g. insects) other than logging appear:to have largely been ignored. The broadening scope of forest resource management necessitates that the basic objective of pest control measures be one of minimising damage or losses not necessarily to one, but several, resources in the ecosystem. Because control measures have beneficial.and detrimental externalities, we need to think in terms of the whole system. The resources considered are probably variable in space and time, and they are not of equal importance. Any.attempt to evaluate impacts in a forest ecosystem necessarily.involves investigating parameters closely associated.with the resources of interest. The pest management strategy should take into account, among other things, the extent of and variation in the resources and impacts of the damage. On poor.and low sites, control of the Douglas-fir tussock moth would be justified if the objective was to decimate populations from potential epicenters. But contagious spread in this insect is insignificant (Mason and Thompson, 1971; Mason, 1974; Wickman et al., 1973). If (passive) contagious spread was significant, it would likely be for long distances. Presence of the insect in major valleys in the area, where cross valley and north<—^south winds are common suggests some populations are probably related, but in a way which is difficult to predict, from dispersal patterns. So far the evidence indicates that patches so.common .on the landscape represent different native or autochthonus populations. Controlling the foci is sound only to the - 129 -extent that it prevents damage of resources within the infestation patch. The "nip-in-the-bud" concept (Stehr, 1968) is not relevant beyond the patch because decimating the insects in one patch does not affect those inv,another. The three year cycle may describe adequately individual populations, but in the field often populations are not in phase with each other. On the landscape longer cycles are not uncommon. A broadcast control treatment is inefficient unless it is applied only on the good and medium sites and when most populations are in the same vulnerable phase, before eruption or crisis. As populations are "independent" and infestations last for a few years, control is likely to be spread over as many years. Some stands may have to be treated more than once. High rates of tree mortality are probably due to chronic infestations. It should not be assumed that every tree which has been stripped of its foliage will. die. It is unrealistic to invoke control measures in a stand during the first year of infestation on grounds that we'are preventing trees from dying. Most outbreaks in the past collapsed naturally, although often after ;severeri damage had already beenj inflicted. Most control attempts were undertaken during the outbreak phase. Since most populations are not in phase with each other, claims of successful control.from a single broadcast treatment may be suspect. Likely the populations would have collapsed naturally, as they have many times in the past. Any spurious claim of successful control is dangerous from at least two angles: it justifies expenditures unwisely spent, and sets a precedent;on,: which future control attempts will be based on the premise that "it worked the last time". - 130 -A decision .which advocates no control in some areas requires less cash outlay. It also.realizes benefits in forage yields. On south and east facing slopes providing winter refuge for mule deer, the decision assumes special dimension. High increases in forage yields should be expected, on mesic sites with deep sandy loam soils. Such sites may produce high timber values but control should not necessarily be the rule.there, especially when the stands are managed primarily for forage, or wildlife refuge. Conclusion By reducing stand stocking and density, the Douglas-fir tussock moth increases range forage production. Significant gains should be expected on mesic sites, in more intensively chronically infested stands, and in dead patches of trees. Salvaging infested stands increases the benefits, especially when followed by range seeding. Tree survival is affected.by defoliation, and is related to size and possibly other factors. Effects on tree radial growth at breast height.are not significant during the first year following defoliation. I did not investigate effects beyond the first year.-It is true that various degrees of contagious spread of tussock moth populations by wind dispersal of the larvae are inevitable. Nevertheless.the strong possibility remains that existing widespread populations have their own independent capability of erupting under the blanket conditions of weather systems which favor population - 131 -increase generally. Delayed eruptions in some localities do not necessarily signify.that, they originated from "hot spots", but may signify that the microsite conditions, were somewhat less favorable, and thus created a lag. Advocating a broadcast control of the pest is unwise because it leads to .spending money on low quality sites and on sites which are managed primarily for forage production. It also prevents realization of range benefits. Suppression measures.may not. prevent remote stands from being.infested because contagious spread.is not common. Selective control, with better sites given priority., will improve efficiency of investing scarce funds. There is' need for undertaking similar, studies in other affected localities to obtain data.from.a wider base more representa tive of the Interior Douglas-fir Dry Subzone. Because outbreaks are recurrent, it is relevant to study impacts in a long term mission oriented research. But we need to know the specific history of stands, and possibly individual trees, if we are to state long term impacts confidently. There is also need to examine changes in forage yield and quality at the species level to provide most meaningful data to range and wildlife-managers and graziers. Possible confounding of yield data may be avoided by using exclosures. - 132 -LITERATURE CITED Anonymous. 1956. Argument of crown counsel (to Sloan Royal Commission on forest resources in British Columbia), section 199. Balch, R.E. 1932. The fir tussock moth (Hemerocampa pseudotsugata McD). J. Econ. Entomol. 25: 1143-1148. Beckwith, R.C. 1976. Influence of host foliage on the Douglas-fir tussock moth. Environ. Entomol. 5(1): 73-77. Beil, C.E. 1974. Forest associations of the Southern Cariboo Zone, British Columbia. Syesis 7: 201-233. Bell, E., R. Fight and R. Randall. 1975. Ace - The two-edged sword. J. For. 73.:. 642-643:^. Berland, L. 1935. Premier resultats des mes recherches en avion sur la fauna et 1$ flora atmospherique. Ann. Soc. Entomol. Fr. 104: 73-96. Blair, R.M. and L.E. Brunett. 1976. Phytosociological. changes after timber harvest in a southern pine ecosystem. Ecology 57(1): 18-32. Bonner, G.M. 1967. Estimation of ground.canopy density from ground measurements. J. For. 65: 544-547. . 1968. A comparison of photo and ground measurements of canopy density. For. Chron. 43 (3): 12-16. Brayshaw, T.C. 1965. The dry forest of southern British Columbia. Ecology of Western N. America (ed. V.J. Krajina). 1: 65-78. Brown, H.E. and D.P. Worley. 1965. Some applications of the canopy camera in forestry. J. For. 63: 674-680. Campbell, R.S. and J.T. Cassady. 1949. Determining forage weights on southern forest ranges. J. Range Manage:-.".'<.,'. 2(1): 30-32. Carolin, V.M. and W.K. Coulter. 1975. Comparison of western spruce budworm populations and damage on grand fir and Douglas-fir trees. U.S. For. Serv. Res. Pap. PNW-195, 16 p. Cole, W.E. 1966. Effect of pine butterfly defoliation on Ponderosa pine in southern Idaho. U.S. For. Serv. Resr Note INT-46, 7p. - 133 -Commonwealth of Massachusetts.. 1908. Suppressing the Gypsy and Brown-tail moths. 3rd Annu. Rep. Comm. Mass. Public Doc. No. 73. p. 162. Condrashoff, S.F. and J. Grant. 1962. Sampling Douglas-fir tussock moth populations. Can. For. Serv. Bi-Mon. Res. Notes 18(4):3. Curtis, D.J. 1974. Douglas-fir tussock moth benefit/cost analysis. Proc. 25th Annu. West. For. Insect Work Conference. Rocky Mtn. For & Range. Exp. Stn., Ft. Collins. Colorado, pp. 42-47. Dahlsten, D.L., E.A. Cameron and W.A. Cooper. 1970. Distribution and parasitization of cocoons of the Douglas-fir tussock moth, Hemeveampa pseudotsugata (Lepidoptera: Lymantriidae), in an isolated infestation. Can. Entomol. 102: 175-181. Dilworth, J.R. and J.F. Bell. 1972. Variable probability sampling -variable plot and three-p. Oregon State University. 129 p. Dodd, C.J.H. 1969. The relationship of herbaceous understory growth to tree overstory. B.S.A. thesis, unpubl. (U.B.C.). 75 p. Dodd,-- CvJ;Hr,..!A'.,-McLean and V.C. Brink. 1972. Grazing values as - related to tree crown covers. Can. J. For. Res. 2(1): 185-189. Dolph, R.E. 1973. Douglas-fir tussock moth on Douglas-fir and Grand fir in the Pacific Northwest region. Proc. 24th West. For. Insect Work Conference. Tucson, Arizona, pp. 33-37. Duff, G.H. and N.J. Nolan. 1953. Growth and morphogenesis in the Canadian forest species. 1. The controls of cambial and apical activity in Pinus resinosa Ait. Can. J. Bot. 31: 471-513. Eaton, C.B. and G.R. Struble. 1957. The Douglas-fir tussock moth in California. Pan-Pac. Entomol. 33(3): 105-108. Eddleman, L. and A. McLean. 1969. Herbage - its production and use within the coniferous forest. Proc. Coniferous forests of the northern Rocky Mountains, pp. 179-196. Francis, R.H., C.E. Smith and J.E. Firth. 1972. Freezing decayed wood to facilitate ring counts and width measurements. U.S. For. Serv. Res. Note PNW-187, 4 p. Freese, F. 1962. Elementary forest sampling. U.S. For. Serv. Agric. Handbook No. 232, 91 p. Frischknecht, N.C. and A.P. Plummer. 1949. A simplified technique for determining herbage production on range and pasture land. Agron. J. 41 (1): 63-65. - 134 -Gaines, E.M., R.S. Campbell and J.J. Brassington. 1954. Forage production of longleaf pine lands of southern Alabama. Ecology 35: 59-62. Garrison, G.A. 1949. Uses and modifications for the "moosehorn" crown closure estimator. J. For. 47:733-735. Gilmer, P.M. 1923. The poison'and.poison,'apparatus of the White-marked tussock moth, Hemerocampa leucostigma Smith and Abbott. J. Parasitol. 10: 80-86. Gordon, D.T. 1962. Growth response of eastside pine poles to removal of low vegetation. U.S. For. Serv. PSW. For & Range Exp. Stn. Res. Note 209, 3p. Graham, D.A. 1974. Insect infestations: the tussock moth. Background and control project. Proc. 53rd Annu. Wash. State For. Conference, Seattle, Wa. pp. 79-89. Graham, K. 1963. Concepts of forest entomology. Reinhold. pp. 176-181. Grant, G.G., D. Freeh and D. Grisdal. 1975. Tussock moths: pheromone cross stimulation, calling behaviour and effect of hybridiza tion (Lepidoptera: Lymantriidae). Ann. Entomol. Soc. Am. 68(3): 519-524. Greenbank, D.O. 1963. The development of the outbreak, in The Dynamics of epidemic Spruce budworm populations (ed. R.F. Morris). Mem. Entomol. Soc. Can. 31: 19-23. Harwood, R.F. 1975. Economics, aesthetics, environment and entomolo^ gists: the tussock moth dilemma. Environ. Entomol. 4(2): 171-174. Hughes, I.G. 1975. Changing attitudes and habitat preferences of two species of wood-ant (Formica rufa and F. lugubris) in North Wales and Salop. Trans. R. Entomol. Soc. Lond. 127(3): 227-239. Hughes, K.M. and R.B. Addison. 1970. Two nuclear polyhedrosis viruses of the Douglas-fir tussock moth. J. Invertebr. Pathol. 16: 196-204. Jameson, A.D. 1967. The relationship of tree overstory and herbaceous understory vegetation. J. Range Manage. 20'(4): 247-249. Koerber, T.W. and B.E. Wickman. 1970. Use of tree ring measurements to evaluate impact of tree defoliation. Proc. Tree ring analysis with special reference to Northwest America (ed. J.H.G. Smith and J. Worrall). Vancouver, B.C. pp. 101-106. - 135 -Kozlowski, T.T. 1969. Tree physiology and., forest pests. J. For. 67: 118-123. Krajina, V.J. 1959. Biogeoclimatic zones in British Columbia. Univ. of B.C. Bot. Series. No. 1, 47p. . 1965. Biogeoclimatic zones and biocoenoses of British Columbia. Ecology of West. N. America 1: 1-17. Leonard, D.E. 1970. Intrinsic factors causing qualitative changes in populations of Porthetria dispar (Lepidoptera: Lymantriidae). Can. Entomol. 102: 239-249. Livingston, R.L. and S. Tunnock. 1973. Biological evaluation of existing Douglas-fir tussock moth populations in northern Idaho to determine damage potential for 1973. Idaho Dept. Lands, Coeur d'Alene. Idaho. Luck, R.F. and D.L. Dahlsten. 1967. Douglas-fir tussock moth '(Hemerocampa pseudotsugata) egg mass distribution on White fir in northeast California.. Can. Entomol..99: 1193-1203. Luhl, von R. and G. Watzek. 1976. Die wirkung von hohlenbrutenden Kleinvb'geln forstinsekten. Allg. Forst. Jagdzig. 147(6/7): 113-116 (English, summary) Martignoni, M.E., P.J. Iwai, K.M. Hughes and R.B. Addison. 1969. A cytoplasmic polydrosis of Hemerocampa pseudotsugata. J. Invertebr. Pathol. 13(1): 15-18. Mason, R.R. 1974. Population change in an outbreak of the Douglas-fir tussock moth, Orgyia pseudotsugata (Lepidoptera: Lymantriidae),. J nf BentFr.aT.-- A-r-.i:zonata:CaiiiktEntomb 1> . 106:_1171-1174. Mason, R.R. and C.G. Thompson. 1971. see * page 139. McConnell, B.R. and J.G. Smith. 1970. Response of understory vegetation to Ponderosa pine thinning in eastern Washington. J. Range Manage. 23(5): 208-212. McDunnough, J. 1921. New British Columbia tussock moth, Hemerocampa pseudotsugata. Can. Entomol. 53: 53-56. McGuire, J.R. 1974. Forest environmental protection: challenges in insect and disease management. Proc. Symposium on the Spruce budworm. Nov. 11-14. Alexandria, Virginia, U.S.A. pp. 3-8. . 1976. Cooperative forest insect and disease management. What's the role of the federal government? South. Lumberman 233(2895): 54-56. McLean, A. 1969. Plant communities of the Similkameen Valley, British Columbia, and their relationships to soils. PhD*thesis, unpubl. Wash. State Univ. 133 p. - 136 -McLean, A., T.M. Lord and A.J. Green. 1970. Utilisation of the major plant communities in the Simllkameen Valley, British Columbia. J. Range Manage. 24(5); 346-351. McLean, A, and J.H.G. Smith. 1973. Effects of climate on forage yields and tree ring widths in British Columbia. J. Range Manage. 26(6): 416-419. Michalson, E.L. 1974. Benefit-cost analysis applied to pest management strategies. Proc. 25th Annu. West. For. Insect Work Conference. Rocky Mtn. For. & Range Exp. Stn. Fort Collins, Colorado, pp. 32-40. Mitchell, K.J. 1975. Dynamics and simulated yields of Douglas-fir. For. Sci. Monogr. 17, 39 p. Morris, R.C. 1976. Silent forests of the Lower Alabama. Ala. For. Products 19(7): 51, 53. Morris, R.F. 1963. Predictive population equations based on key factors. Mem. Entomol. Soc. Can. 32: 16-21. Munro, J.A. and I. McT. Cowan. 1947. A review of the bird fauna of British Columbia. B.C. Prov. Museum Special Public. No. 2, 285 p. O'neal, L.C. 1962. Some effects of artificial defoliation on the growth of Jack pine (Pinus banksiana Lamb). Can. J. Bot. 40: 273-280. . 1963. The suppression of growth rings in Jack pine in relation to defoliation by the Swaine Jack pine sawfly. Can. J. Bot. 41: 227-235. Pase, CP. 1958. Herbage production under immature Ponderosa pine stands.in the Black Hills. J. Range Manage. 11(5): 238-243. Pattullo, T.D. 1920. Forests and forestry in British Columbia. Special Report to the Imperial Forestry Conference. London. 35 p. Pearse, P.H. 1976. Timber rights and forest policy in British Columbia. Report of the Royal Commission of forest resources 1. p. 205. Peddie, W. 1938. The establishment of the trichomatic theory of color vision. Proc. R. Soc. Edinb. 59: 15-21. Polge, H. and S. Garros. 1971. Influence de defoliaisons sur la structure du bois de pin maritime. Ann. Sci. For. (Paris) 28(2): 195-206 (English summary). - 137 -Rivas, A.M. 1974. Economic evaluation of mountain pine beetle in the Targhee National Forest. Proc. 25th Annu. West. For. Insect Work Conference. Rocky Mtn. For and.Range Exp. Stn. Fort Collins, Colorado, pp. 40-42. Robinson, M.W. 1947. An instrument to measure forest crown cover. For. Chron. 23(3): 222-225. Rowe, J.S. 1972. Forest regions of Canada. Can. For. Serv. Publ. No. 1130, 171 p. Schweitzer, D.L., R.W. Sassaman and CH. Schalla\i. 1972. Allowable cut effect - some physical and economic implications. J. For. 70(7): 415-418. Severson, K.E. and J.J. Kranz. 1976. Understory production not predictable from aspen basal area or density. U.S. For. Serv. Res. Note RM-314, 4 p. Silver, G.T. 1962.. The distribution of Douglas-fir foliage by age. For. Chron. 38(4): 433-438. Sloan, G. McG. 1945. Report of the Commissioner relating to the forest resources of British Columbia. Victoria, B.C. 195 p. . 1956. The forest resources of British Columbia: report of the Commissioner on forest resources. 2. pp. 417-818. Smith, J.H.G. 1970. Persistence, size and weight of needles on Douglas-fir and Western hemlock branches. Can. J. For. Res. 2(1): 173-178. . 1974. Dynamics of stand development as related to wild life. Reprint from Wildlife and forest management in the Pacific Northwest (ed. H.C. Black) Oregon State Univ. 14 p. Solomon, R.M., P.F. Ffolliot and J.R. Thompson. 1976. Correlation between transmissivity and basal area in Arizona Ponderosa pine forests. U.S. For. Serv. Res. Note RM - 318, 3 p. Staebler, G.R. 1963. Growth along the stems of fully-crowned Douglas-fir trees after pruning to specified heights. J. For. 61: 124-127. . 1964. Height and diameter growth for four years following pruning of Douglas-fir. J. For. 62: 406. Stark, R.W. 1975. Forest insect pest management, in Introduction to insect pest management (ed. R.L. Metcalf and W.H. Luckman). J. Wiley & Sons. pp. 509-528. - 138 -Stark, R..W-. , :and J.A. Cook. 1957. The effects of defoliation by the Lodgepole needle miner. For. Sci. 3: 376-396. Stehr, G. 1968. On some concepts on the population biology of the Spruce budworm. Proc. Entomol. Soc. Ont. 99:54-56. Stein, W.I. 1955. Pruning to different heights in young Douglas-fir. J. For. 53: 352-355. Sugden, B.A. 1957. A brief history of outbreaks of the Douglas-fir tussock moth, Hemerooampa pseudotsugata McD., in British Columbia. Proc. Entomol. Soc. B.C. 54: 37-39. Thomas, J.W. and D.C. McCluskey. 1974. Effects of aerial application of DDT for tussock moth control on nestling survival of mountain bluebirds and house wrens. U.S. For. Serv. Res. Pap. PNW-185, 37 p. Tisdal, E.W. and A. McLean. 1957. The Douglas-fir zone of southern British Columbia. Ecol. Monogr. 27: 247-266. Tunnock, S. 1973. The Douglas-fir tussock moth in the Northern Region - a cartographic history of outbreaks from 1928 to 1973. U.S. For. Serv. Div. State & Priv. For. North. Reg. Report •No. 73-27, 18 p. Turcek, F.J. 1948. Birds'in an oak forest during a Gypsymoth outbreak in South-Slovakia. Am. Midi. Nat. 40(2): 391-394. U.S.D.A. For. Serv. 1973a. Pac. Northwest For. Range Exp. Stn. Ann. Rep. 1972, 52 p. . 1973b. USDA-USDI Environmental statement cooperative Douglas-fir tussock moth pest management plan. Idaho-Oregon-Washington. Pac. Northwest Region. Portland, Oregon. 301 p. + appendix. Van Dyne, G.M., W.G. Vogel and H.G. Fisser. 1963. Influence of small plot size and shape on range herbage production estimates. Ecology 44 (4): 746-759. Vincent, A.B. 1962. Development of Balsam fir thickets in the Green River watershed following the spruce budworm outbreak of 1913-1919. Can. Dept. For., For. Res. Br. (Ottawa), Tech. Note. No. 119, 11 p. Webb, W.L. and K. Kilpatrick. 1976. Defoliation of Douglas-fir in tussock moth outbreak near Kamloops, B.C. Proc. Symp. Terrestrial & Aquatic Studies of the Northwest (ed. R.D. Andrews et al.). E. Wash. State Coll. Press., Cheney, Wash, pp. 135-143. - 139 -Wellington, W.G. 1945. Conditions governing the distribution of insects in the free atmosphere. IT. Can. Entomol. 77: 21-28. . 1948. The light reactions of the spruce budworm, Ckoristoneura fumiferana. Can. Entomol. 80: 56-82. Wickman, B.E. 1958. Mortality of White fir following defoliation by the Douglas-fir tussock moth in California.. 1957. U.S. For. Serv. Cal. For. Range Exp. Stn. Res. Note 137, 4 p. . 1963. Mortality and growth reduction of White fir following defoliation by the Douglas-fir tussock moth. U.S. For. Serv. Res. Pap. PSW-7, 15 p. Wickman, -B.E. ,V R.R. Mason and C.G. Thompson. 1973. Major outbreaks of the Douglas-fir tussock moth in Oregon and California. U.S. For. Serv. Techn. Rep. PNW-5, 18 p. Wickman, B.E., and D.A. Renton. 1975. Evaluating damage caused to a campground by Douglas-fir tussock moth. U.S. For. Serv. Res. Note PNW-257, 6 p. Wickman, B.E., and R.F. Scharpf. 1972. Decay in White fir top-killed by Douglas-fir tussock moth. U.S. For. Serv. Res. Pap. PNW-133, 9 p. Williams, C.B. 1967. Spruce budworm damage symptoms related to radial growth of Grand fir, Douglas-fir, and Engelmann spruce. For. Sci. 13: 274-285. Williams, R.G. 1974. Considerations in benefit-cost analysis. Proc. 25th Annu. West. For. Insect Work Conference. Rocky Mtn. For. Range Exp. Stn. Fort Collins, Colorado, pp. 26-32. Wolfenbarger, D.O. 1946. Dispersion of small organisms. Am. Midi. Nat. 35(1): 1-152. Wright, K. 1974. Insect infestations: the tussock moth. Future ^approaches to forest pest control. Proc. 53rd Annu. Wash. State For. Conference, Seattle, pp. 101-107. Zanforlin, M. 1970. The inhibition of light orientating reactions in catterpillars of Lymantriidae3 Lymantria dispar and Orgyia antiqua (Lepidoptera). Monit. Zool. Ital. 4(1): 1-19. Mason, R.R. and C.G. Thompson. 1971. of the Douglas-fir tussock (Lepidoptera:Lymantrii dae) Collapse of an outbreak population moth '.Hemerocampa pseudotsugata U.S. For. Serv. Res. Note PNW-139, 10p. - 140 -i Appendix - 141-SCIENTIFIC. NAMES OF PLANTS Trees Balsam fir Black cottonwood Black spruce Black tupelo Choke Cherry Abies balsamea (L.) Mill. Populus trichocarpa Torr. & Gray Picea mariana (Mill.) BSP Nyssa sylvatiaa Marsh. Prunus virginiana var. melanocavpa (A. Nels.) Sarg. Douglas-fir (interior) Pseudotsuga menziesii (Mirb.) Franco var. glauoa Douglas maple Engelmann spruce Grand fir Jeffrey pine Longleaf pine Ponderosa pine Subalpine fir Western hemlock Western larch White fir Acer globurum Torr. var.douglasii (Hook) Dipp. Picea engelmannii Parry Abies grandis (Dougl.) Lindl. Pinus geffreyi Grev. & Balf. Pinus palustris Mill. Pinus ponderosa Laws. Abies lasiooarpa (Hook) Nutt. Tsuga heterophylla (Raf.) Sarg. Larix occidentalis Nutt. Abies conoolor Gord. & Glend. Understory vegetation Balsam root Balsamrhiza sagitata Nutt. Blue bunch wheat-grass Agropyron spicatum (Pursh) Scribn. & Smith. Downy brome Bromus tectorium L. June grass Koeleria cristata (L.) Pers. - 142 -Kentuky bluegrass Needle-and-thread Pine grass Sagebrush Timber milk-vetch Poa pratensis L. Stypa Qomata Trin. & Rupr. Calamagrostis rubesaens Buckl. Artemesia tridentata Nutt. Astragalus miser Dougl. ex Hook I NATIONAL TOPOGRAPHIC SYSTEM BRITISH COLUMBIA SECOND STATUS EDITION—July 15th, 1968. SHEET 92 VNE Surveyed Crown Land in flooded areas Surveyed land in flooded areas which has been alienated Surveyed Timber Lease. Licence, or Berth Induin Reserve Government Reserve Land District Boundary Provincial Forest Boundary Tree Farm L Municipality Water Supply Area Park Park less than 10 acres Campground Forest Service Lookout Post Office School Church Hospital Building Cemetery Mine Dyke Historic Monument On the above index,maps published with district lot land status are coloured yellow. Land Commissioner's OJJice is located at Kamloops. Mineral Claims arc not shown on this sheet. District land lot numbers 24 Sections within Townships 24 KAMLOOPS LAKE, B. C SHEET 921/NE SECOND STATUS EDITION 

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.831.1-0075399/manifest

Comment

Related Items