Home

Estimating the Abundance of Arboreal Forage Lichens: User's Guide

Contents

1. 8 2 15 55 55 11 0 20 17 7 8 8 26 5 25 53 10 6 15 9 30 0 9 1 8 1 8 4 5 35 6 1 12 2 18 3 40 5 1 10 2 5 1 20 4 45 5 9 5 9 11 8 50 2 1 6 3 4 2 12 6 55 7 6 7 6 60 5 7 5 7 65 1 9 1 9 3 8 70 1 3 1 3 dD 0 7 0 7 80 0 3 0 3 Live trees ha 7 6 4B 65 6 30 5 19 148 6 Snags ha 71 7 1 Calculatinglichen biomass from stand tables To producealichen biomasstable arrange the Lichen Class data by tree species and diameter class The approximate biomass midpoints of the Lichen Classes provide an estimate of the amount of lichen in grams on atreein a given Lichen Class Biomass midpoints are used as multi plier factors to calculatethe amount of lichen contributed by thetreesin a diameter class The Lichen Class conversion factors are Lichen Class M ultiplier 2 5 25 150 450 800 OBWNrF As Lichen Class 5 has no midpoint 800 is suggested as a multiplier It bears the same proportional relationship to the Class 4 5 boundary as the Class 4 midpoint does to the Class 3 4 boundary Table 4 is a lichen biomass table based on the data in Tables 2 and 3 A species summary Table 5 shows the contribution of each tree species and of dead trees to the overall lichen biomass in the proposed block Be aware that the biomass values produced in this way
2. field cards electronic spreadsheet Specify type electronic database Specify type cruise compilation other Specify If requested would you release your data to a central database Credit will be given if data are used Comments Please mail or fax to Research Wildlife Habitat Ecologist or Wildlife Section Head Ministry of Forests Ministry of Environment Lands and Parks Cariboo Forest Region Omineca Peace Region 200 640 Borland Street 1011 Fourth Avenue Williams Lake BC V2G 4T1 Prince George BC V2L 3H9 Fax 250 398 4406 Fax 250 565 6629 35
3. 59 47 1 7 065 0 6 9 0 0 6 9 3 105 0 Ki 0 0 0 0 0 0 10 637 5 72 7 25 0 0 0 0 0 0 0 0 5 0 45 9 5 237 5 22 6 9 0 316 4 740 0 12 6 0 0 12 6 5 670 0 2 5 0 0 2 5 2 000 0 12 647 5 56 2 30 0 0 0 0 0 0 0 0 0 0 0 8 0 8 20 0 216 15 23 1 3 465 0 7 9 15 9 4 4 230 0 0 0 0 0 0 0 0 0 7 715 0 33 3 35 0 0 0 0 0 0 0 0 3 8 0 0 3 8 95 0 9 5 42 13 7 2 055 0 5 7 8 4 14 1 6 345 0 14 0 0 14 1 120 0 9 615 0 33 0 40 0 0 0 0 0 0 0 0 0 0 2 6 2 6 65 0 5 0 5 2 10 2 1 530 0 10 1 2 6 12 7 5 715 0 0 0 0 0 0 0 0 7 310 0 25 5 45 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 8 6 3 2 118 1 770 0 E 3 2 7 5 3 375 0 al 0 0 0 0 0 0 5 145 0 19 3 50 15 1 3 2 8 7 0 0 0 0 0 0 0 0 0 0 0 3 8 3 8 570 0 6 1 2 5 8 6 3 870 0 0 0 0 0 0 0 0 0 4 447 0 15 2 55 0 0 0 0 0 0 0 0 15 0 0 15 37 5 3 0 7 6 10 6 1 590 0 B 0 0 0 3 0 1 350 0 0 0 0 0 0 0 0 0 2 977 5 15 1 60 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 5 7 6 8 1 020 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 020 0 6 8 65 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 1 9 3 9 585 0 0 0 0 0 0 0 0 0 0 0 1 9 1 9 1 520 0 2 105 0 5 8 70 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 13 13 585 0 0 0 0 0 0 0 0 0 85 0 13 75 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 7 0 7 105 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 105 0 0 7 80 0 0 0 0 0 0 0 0 0 0 0 3 0 3 75 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 75 0 3 Total 561 68 9 1573 1177 37 1514 37850 1 3 487 2450 37500 67 195 812 355400 39 19 58 46400 818723 5463 stems ha Lichen Class Multiplier 1 2 5 2 25 3 150 4 450 5 800 Tables 6 and 7 show th
4. Board and should ensurethat work areas will besafe 280 Legend HE Pre harvest 240 200 Post harvest 160 120 E P 0 a 1 2 3 4 5 Lichen Class Trees per hectare figure 12 Trees per hectare in Lichen Classes 1 5 before and after a single tree selection harvest table 7 Post harvest summary by species based on Table 6 Lichen Subalpine Class fir Spruce Snags Total 1 stems 56 7 0 63 grams 140 17 0 157 2 stems 118 34 0 151 grams 2 943 842 0 3 785 3 stems 196 49 0 245 grams 29 445 7 305 0 36 750 4 stems 62 20 0 81 grams 27 765 8 775 0 36 540 5 stems 4 2 0 6 grams 3 120 1 520 0 4 640 Total gramyha 63 413 18 460 0 81872 grams TI 23 0 100 Total stems ha 436 111 0 546 Values may not sum to totals because of rounding 30 Users can help improve adaptive management in caribou habitat by contributing their data to a central registry and by monitoring lichen abundance in selection harvesting blocks 7 USING THE HELD GUIDE TO IMPROVE ADAPTIVE MANAGEMENT What is adaptive management Adaptive management is a structured process for learning from the results of operational activities so that future management can beim proved Thereismuch to belearned about the potential of uneven aged silviculture systemsto maintain caribou habitat M onitoring the results of
5. a stand management pre scription see Section 6 Chi square tests are familiar and easy to use G tests are less common but just as easy to calculate Both tests can also be applied to summarized data such as the number of high or low scores if the full array of scores is unavailable The chi square method has an important limitation however Small score frequencies at the upper or lower tail of the distri bution may distort the X statistic and give unreliable results When one or more frequencies is lt 5 the usual advice isto combine them with the next adjoining class e g Lichen Class 1 with Class 2 or Class 5 with Class 4 However the number of high or low scores may bethe factor that differs most between onesample and another and if score classes are combined this information is lost TheG test is considered morerobust to this limitation than the chi square test For a G test frequencies lt 5 do not need to be combined especially if G is adjusted as G q For this reason the G test isthe better choice The most powerful test for use with either oneor two score frequency distributions is the Kolmogorov Smirnov K S test For aK S test unlike chi square small frequencies at the tails do not have to be com bined indeed it is disadvantageous to do so sincetheK S testis sensitive to any difference in asymmetry of samples that are similar in other respects Unfortunately K S tests are restricted to just one sample o
6. an infinite range of possible values that could be analyzed using parametric ANOVA or Student s t test The appearance of continuity is misleading however since each score class spans a range of lichen abundance but has only a single biomass conversion factor The range of possible values in any onesampleis restricted rather than infinite and its distribution is disjunct rather than continuous Therefore lichen data expressed in biomass equivalents are best analyzed by non parametric methods The Kruskal Wallis test for independent samples and the Friedman test for related samples are suitable non parametric tests for comparing lichen biomass at the plot level 19 A remark on theuse of statistical tests Statistical tests area helpful analytical tool but final judgement rests with the user not on the test alone Other considerations should be taken into account For example perhaps the samples are small but do they differ distinctly nonetheless asin Figures 6 and 7 H ow bigistherisk that the differenceisdueto chance By convention in statistical tests a risk greater than 5 i e p gt 05 isconsidered high But if therisk isonly 6 or 7 or even 10 a greater likelihood still existsthat the differenceis genuine Lastly differences at the higher Lichen Classes may be biologically signifi cant even if not statistically significant and they may also differ spatially even if notin other respects Theinternal structure of thesa
7. and 3 o ecesssessssessssssesssesssseseseseessstsssssseesssssssssesssetusseteansesene 29 7 Post harvest summary by species based ON Table 6 o cccccccsssscsssssssesssssssessusssseessssssessssssseeseesssstssesssssteenssssseteve 30 FIGURES 1 Range of caribou in British Columbia on ccessssesssssessusssseescussssssecssssesssssussesesssusssasssusssseessssssnssssussanesecsinsesseeien 4 2 How to use the scale to adjust lichen scores wceccssessssssssssussssesssussseesssssssesssessssessssusssotesssssseeissssseeessssseesesessseeeeeses 6 3 Frequency distribution of the difference between Lichen Class based on measured samples and Lichen Class assessed by an observer oo eesesssssssessesssssusssssnusstnnsssssnsssssasstiensssse 8 4 Frequency distribution of Alectoria and Bryoria biomass in the lower CANOPY of Trees iM AN ESSF stand eccsccsssesssssssssseessnssssssseessusssssassesesusssssssssesesssusssetesssssssesssuessesussseaett 10 5 Frequency distribution of Lichen Class scores based on the same CALA EE AS FIgUrE Miann ar a ea AANE OONN RE ee R 10 6 Lichen score frequencies trees gt 7 5 cm dbh at two sites used as MOUN EAM Car DOU winter MAN GE soeren a a a hakina ul z Lichen score frequencies trees gt 7 5 cm dbh at two sites not used AS MOUNTAIN caribou winter range o cacccssssscsssssssssssssssssussssesssssssssssssssoesissssseessssssessssssseeseessusessessssssttesesssetessssssseeessset u 8 Lichen plots at random distances along two transe
8. distribution of Lichen Class scores based on the same dataset as Figure 4 10 following type hererepresenting two hypothetical sitesto be Lichen Classscores 1 2 3 4 5 Total compared A frequency distribution table is a better summary of lichen 0 data than the mean score Site 1 no of trees 1 Site 2 no of trees 0 13 23 31 19 1 88 11 20 29 20 2 82 Bar graphs based on frequency distributions are useful tools for assessing whether sites are similar or different and whether they have high or low Legend Site 1 Site 2 Percent of scores 2 3 4 5 Lichen Class figure 6 Lichen score frequencies trees gt 7 5 cm dbh at two sites used as mountain caribou winter range Legend N E sites W sites N Percent of scores W D e gt Lichen Class figure 7 Lichen score frequencies trees gt 7 5 cm dbh at two sites not used as mountain caribou winter range levels of lichen abundance Figure 6 shows the frequency distribution of Lichen Classes at two sites used regularly by mountain caribou in winter Thefrequencies are plotted as percent ages of each sample Both sites have ahigh proportion of treesin the classes thought to be useful to caribou 67 of scores gt 3 at Site 1 and 57 at Site2 In contrast Figure 7 shows the fr
9. lichen than the trees on pages 16 17 of the Field Guide and less lichen than the trees on pages 18 19 W hen you havesome experience you may beableto skip thefirst photo series A C and go directly to the second set of photos H owever to prevent your estimates from wandering itisimportant to confirm your assessment by referring to the photos Even experienced users should refer to the photo guide several times a day and more frequently where lichen abundance is especially variable Questionsand answers Can lichen assessments be done at any time of year Lichen assessments are best done during the snowfree season As the 4 5 m point is always located relative to the forest floor it is easiest to identify when thereis no snowpack Also assessments can be difficult when fresh snow has accumulated on branches Some of these trees have hardly any lichens on them Do I have to assess them Yes The aim is to find out how many trees are present in each Lichen Class not just to look for trees with a lot of lichen To avoid biasing the results each tree that is identified as a sample tree must be assessed regardless of how much lichen ison it If you are assessing trees with very little lichen remember that a tree with only awisp of lichen belongsin Class 1 not Class 0 Lichen Class 0 is reserved for trees with no lichen at all This is extremely rare in most stands where lichen assessments are done Class 0 is included in the s
10. of Lichen Classes that existed before harvesting while removing no more than 30 of the timber volume Ideally if the har vesting can befocused on low scoring trees the proportion of trees with high lichen abundancecan beincreased 12 Stevenson S K H M Armleder M J Jull D G King E L Terry G S Watts B N McLellan and K N Child 1994 M ountain Caribou in M anaged Forests Preliminary Recommendations for M anagers B C Ministry of Forests Victoria B C Lichen abundance data can help with e planning locations and boundaries of treatment units e projecting the impacts of potential prescriptions on lichen abundance e planning how dead trees will be managed 28 67 table6 Lichen biomass summary table for a single tree selection prescription for subalpine fir B and spruce S based on Tables 2 and 3 Lichen Class 1 Lichen Class 2 Lichen Class 3 Lichen Class 4 Lichen Class 5 All Lichen Classes dbh B S total total B S total total B S total total B S total total B S total total total total class s ha s ha s ha gha shar s ha s ha gha shar s ha s ha g ha s hat s ha s ha gha s ha s ha sha gha gha s ha 10 49 5 0 0 495 1238 70 1 8 2 78 3 1 957 5 20 9 0 0 209 3 135 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 216 3 1487 15 5 1 5 5 10 6 26 5 30 4 55 35 9 897 5 608 0 0 60 8 9 120 0 5 1 0 0 5 1 2 295 0 0 0 0 0 0 0 0 12 339 0 112 4 20 0 0 0 0 0 0 0 0 6 9 11 8 18 7 467 5 12
11. thisvariation at thesitein question When each sitediffers use of a standard sampling plan with a pre ordained layout samplesize and sampling method transect or plot isunlikely to achievethis Existing datasets have shown that much of the sample variation at any particular site is related to differences in spatial scale Even when the site in general isrich in lichens abundance often varies greatly at alocal level Differences from treeto tree within asingleplot may beaslargeas between neighbouring plots and not much lessthan the variation over thesiteasa whole Thus if the study is to include broad scale factors such as elevation or forest cover type a total sample well in excess of 150 trees may be required if the role of these factors is to be detectable among local effects The same may be true for the converse aim of detecting local effects e g site class or light regime if the sample has to be spread over a wide area to obtain enough trees in each category of interest At sites that are poor in lichens on the other hand plots tend to be more internally homogeneous neighbouring plots more often resemble oneanother and most of the variation is at a broader scale Thus fewer sample trees or plots may be needed per unit area the more important factor will betheir dispersion Therefore whenever practicable a pilot survey of the site is advisable see Section 5 to gauge the level of varia tion at different spatial scales b
12. yourself so that you can see as much of the lichen as possible and score the tree on the basis of what you see If a substantial amount of lichen cannot be seen from that position you may need to adjust the class upward Suppose for example you are assessing a tree that looks like a high Class 3 but your estimate does not take into account some branches that were invisible from your viewing position You estimate that those branches support about one third as much lichen as the portion of the tree you have already estimated Use the scale on pages 4 5 of the Field Guide to determine whether the additional lichen will move your estimate to Class 4 Figure 2 Sometimes it is necessary to make allowances for a tree with an asymmetrical crown Again usethescaleon pages 4 5 of theField Guideto maketheadjustment How important is it to identify the 4 5 m point accurately For most purposes the 4 5 m point can be estimated without an aid If lichen abundanceis being assessed repeatedly on thesametrees it ismoreimportant to locatethe 4 5 m point consistently You can usea height pole or cut a stick that reaches to 4 5 m when held abovethehead or useanother person for reference A 178 cm 511 person standingwith onearm extended upwards hasa reach of about 2 25 m or half the height of thelichen estimation zone If you arenot working with a partner you can position yourself at a horizontal distance of 4 5 m less your eye height from thetree
13. 0 0 0 0 0 0 5 1 5 1 127 5 10 0 10 2 20 2 3 030 0 20 0 5 1 25 111 295 0 0 0 0 0 0 0 0 0 14 452 5 50 4 45 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 16 3 5 9 22 2 3 330 0 8 1 5 9 14 0 6 300 0 0 0 0 0 0 0 0 0 9 630 0 36 2 50 2 5 2 1 46 115 0 0 0 0 0 0 0 0 0 0 6 3 6 3 945 0 10 1 42 14 3 6 435 0 0 0 0 0 0 0 0 0 7 891 5 25 2 55 0 0 0 0 0 0 0 0 1 5 0 0 1 5 37 5 3 0 7 6 10 6 1 590 0 3 0 0 0 3 0 1 350 0 0 0 0 0 0 0 0 0 2 977 5 15 1 60 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 5 7 6 8 1 020 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 020 0 6 8 65 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 1 9 3 9 585 0 0 0 0 0 0 0 0 0 0 0 1 9 1 9 1 520 0 2 105 0 5 8 70 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 13 1 3 585 0 0 0 0 0 0 0 0 0 585 0 1 3 75 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 7 0 7 105 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 105 0 0 7 80 0 0 0 0 0 0 0 0 0 0 0 3 0 3 75 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 7 5 0 3 Total 571 76 647 161 8 123 7 430 166 7 41675 2535 656 3191 47 865 0 891 305 1196 53 8200 50 19 69 55200 1115343 6770 stems ha Lichen Class M ultiplier 1 2 5 2 25 3 150 4 450 5 800 table 5 Pre harvest summary by species based on Tables 2 3 and 4 Lichen Subalpine Class fir Spruce Snags Total 1 stems 57 8 15 80 grams 143 19 38 200 2 stems 124 43 16 183 grams 3 093 1 075 410 4 578 3 stems 254 66 47 367 grams 38 025 9 840 6 975 54 840 4 stems 89 31 8 128 grams 40 095 13 725 3 690 57 510 5 stems 5 2 0 7 grams 4 000 1 520 0 5 520 Total gramyha 85
14. 355 26 179 11 113 122 647 grams 70 21 9 100 Total stems ha 528 149 86 763 Values may not sum to totals because of rounding Frequency distributions For comparison with other stands frequency distributions of lichen classes may begraphed using theinformation in thestand tables Figure 10 Producingamap M apped results of lichen surveys may be useful for a variety of purposes within blocks such as identifying different treatment units laying out group selection openings or locating wildlife tree patches or other long term retention areas Figure 11 is a block map showing the number of trees in each 0 01 ha plot with Lichen Class 3 or greater and how that information could be used to locate openings that will maximize the retention of lichen forage after thefirst harvest entry Of course targeting areas of low lichen abundance for harvest openingsis not always appropri ate That will depend on thelong term management objectives and the reasons for low lichen abundance in certain portions of the block Planning the prescription M any factors affect the development of a silviculture prescription in caribou habitat Some of these are e higher level plan direction economic considerations e pre harvest stand structure e ecological requirements of desired tree species e terrain and soil conditions windthrow hazard e forest health concerns abundance and distribution of lichensin the stand contribution of dead tre
15. A srseisiiiiannmiennens iaai A eee A 2 Why Assess Lichen Abundance oa ceccsssssssssssssssssnsesssssssesssssssusssssssssssesssssssstiassesieestmssesseesssiase 3 Caribou depend on lichens for winter forage sssessssessessesssesssssseseessetussssseesee Information about lichens may be required for approval of plans Using the Field Guide to Assess Individual Trees oooo ccccccccccccssssssssssssssssussssesssssssesssusssesseessssesseesess 5 Basic proc dures ssi en a a shonsseeesustitlsteelusove sabes a 5 Questions and ANSWETS oiceececcccccccccccsssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssssissusssssssassssssssssssseeessessssssssssssesee 6 Basic Characteristics of Lichen D atasets ooo ccccssssssssssssssssssssssssnssssssssssssssssssssssssssssssssssssssssesee 10 Using the Field Guide for Reconnaissance 0 ccssssccssssssssusssssssssussessessusseseessusssasesuusssnsesusssssesenssaseses 12 Reconnaissance objectives for lichen assessment Sampling methodS escsesssssnsssssessssessessnecseresseeesseeaeessee Defining the sampling domain Determining the sample size Laying out the sample s Compiling the data wun Analyzing the data M apping the data oe eeccssssessessessssnessnsseeeneeseeseee Using the data to manage caribou habitat cessssssssssessssesssssesesssusssessssssussssessssssessaessssasseeeese Using the Field Guide for
16. Armleder H M S K Stevenson and S D Walker 1992 Estimating the Abundance of Arboreal Forage Lichens B C Ministry of Forests Land M anagement H andbook Field Guide Insert 7 Victoria B C Meidinger D and J Pojar 1991 Ecosystems of British Columbia B C Ministry of Forests Special Rep Ser No 6 Victoria B C Who isthe handbook for This handbook isintended for peoplewho plan forestry activitiesin caribou habitat At the reconnaissance level it is intended for people who e design fieldwork for plans such as Forest Development Plans Landscape Unit Plans Total Resource Use Plans or Local Resource Use Plans e compileand interpret data for those plans At the block level it is intended for people who e plan timber cruising or data collection for Silviculture Prescriptions e develop Silviculture Prescriptions for selection harvesting blocks in caribou habitat At both levels it is for people who e train and supervise crews that collect resource data It may also be used by researchers studying the distribution and abun dance of arboreal lichens or the impacts of management activities on the lichens Where can it be used The Field Guide was developed for usein the ESSF biogeoclimatic zone There the snowpack is often as much as 3 m deep which allows caribou to reach lichens up to 4 5 m This level was used in the Field Guideas the uppe limit for estimating lichen abundance Caribou use arboreal lic
17. E Reconnaissance objectives for lichen assessment Lichen assessments at the reconnaissance level are usually madefor oneor all of the following objectives e to classify areas into broad categories of lichen abundance e to identify and roughly map discontinuities in lichen abundance e to determinethe dominant lichen genusin an area eg mainly Alectoria mixed Alectoria and Bryoria or mainly Bryoria For operational reconnaissance purposes the Field Guide will nor mally be used in areas where site conditions make stands suitable as winter range for caribou In addition to lichen abundance two of the most important site conditions affecting caribou use are Slope and topography Caribou use areas with moderate topography more than areas with rugged topography They use slopes lt 45 more than steeper slopes Stand structure Caribou seem to prefer stands open enough so they can seearound them Thismay beonereason they avoid denseyoung stands The following recommendations also apply to sampling for other purposes such as management planning at the stand or block level Section 6 and research or inventory Section 8 Sampling methods Various sampling methodscan beused at thereconnaissancelevel Some methods that havegiven comparableresults are e plots at intervals along line transects e selectively located plots within strata predetermined for other purposes e plots atregular spacingson agrid e lineor belt
18. LAND MANAGEMENT HANDBOOK Estimating the Abundance of Arboreal Forage Lichens User s Guide A Guide for Users of Estimating the Abundance of Arboreal Forage Lichens Land Management Handbook Field Guide Insert 7 1998 BRITISH Ministry of Forests COLUMBIA Research Program Estimating the Abundance of Arboreal Forage Lichens User s Guide A Guide for Users of Estimating the Abundance of Arboreal Forage Lichens Land Management Handbook Field Guide Insert 7 Susan K Stevenson Art N Lance and Harold M Armleder dy ini of Forests W ON acho Canadian Cataloguing in Publication Data Stevenson S K Estimating the Abundance of Arboreal Forage Lichens U ser s Guide Land Management Handbook 43 A Guide for Users of Estimating the Abundance of Arboreal Forage Lichens Land Management Handbook Field Guide Insert 7 B C Ministry of Forests Includes bibliographical references p ISBN 0 7726 3549 8 1 Lichens British Columbia 2 Caribou Feeding and feeds British Columbia 3 Caribou Habitat British Columbia 4 Forest management British Columbia I Lance Art N I1 Armleder H M III British Columbia Ministry of Forests Research Branch IV Series QK587 7 B7S73 1998 579 7 C98 960122 6 Prepared by Susan K Stevenson R P Bio Silvifauna Research 101 Burden St Prince George BC v2m 298 Art N Lance R P Bio Industrial Forestry Service Ltd 1595 Fifth Ave Prince George BC v2I 319 Haro
19. Lichen Class Statisti cal methods as described in Section 5 may be used to determine whether the prescription succeeded in maintaining the pre harvest frequency distribution of lichen scores If it did statistical analysis should show no significant difference between frequency distributions before and after harvesting Itisalso important to assess theimpact of the harvesting on overall lichen abundance A simple way to do thisisto comparethenumber of trees per hectare pre and post harvest in Lichen Classes gt 3 Thistype of information will help managersto know what relative changein lichen abundanceto expect from various harvesting prescriptions To expressthe changein absoluteterms lichen scores may beconverted to biomass equivalents Section 6 and plot totals calculated It is also possibleto track the fate of thelichensin aharvested block by identifying the sources of the 31 lichen loss Figure 13 shows how two prescriptions used in an experimental selection harvesting block affected the amount of lichen lost through prescription harvest road and landing rights of way clearing and other sources and theamount remaining after harvesting Lichen remaining Lichen lost from inthe stand the stand i 100 80 60 40 20 i 0 nm E Percent of original lichen biomass Remaining Loss from Loss from Loss from Loss from Loss from on residuals felled snags roads and residuals p
20. Planning at the Block Leva oo ccccccsssssusssssessssssaseusssseeseassneseen Block level objectives for lichen assessment Sampling methodS eescesssssesssssesstseesseseeeseessstsesseseee Compiling the data sites ae nnttivtea cae TAA E EA N a i Planne prESCRDUON ar ireen aa ets a a TEEN Using the Field Guide to Improve Adaptive M anagement 000 ccscccsssssscssssssssssssessessssssteessssssses What is adaptive management ccsscscessssssssssseesesssaeseessusseseesusseeeeen Contributing to the Knowledge base on lichen abundance u s Monitoring the effects of selection harvesting on lichen abundance Research Inventory 9 Conclusion cia cess scccspsctrssesttoccccccacitttinnaenatttctataaannanarsrnsssencntracaanticy annnntthonamaamnanetinishcrenancigiaiinenecttsaa 34 Data Registration FO M smansi A 35 TABLES 1 Frequency distribution of Lichen Class scores with associated information on 16 2 Stand table for subalpine fir by Lichen Class 3 Stand table for spruce by Lichen CASS vv ecccccssssssssssssssssssssssssessssssessssssseeesssssseessussssessssssssesssssusseeeessssssatesessseeeissssssesees 4 Lichen biomass summary for subalpine fir and spruce based ON Tables 2 and 3 aves 26 5 Pre harvest summary by species based ON Tables 2 3 and 4 a eescsesssesssssnssssussssnssssssssessssssssssssussstssssssse 27 6 Lichen biomass summary table for a single tree selection prescription for subalpine fir and Spruce based ON Tables 2
21. and use aclinometer to find the point on thetrunk at a 45 100 angle from your viewing position Some trees have to assess don t look like the trees in the photos How do I assess them Some trees such as snags or pines look very different from the trees in the photographs They may have all the lichens clumped on a few small dead branches and on the trunk To assess such trees it may help to use the 5 gram clumps provided for reference on each tree photo The relationship between Lichen Class and number of 5 gclumpson a treeisshown on page 21 of the Field Guide w It is hard to tell the difference between Classes 1 2 and 3 from the photographs Is there any other way to classify trees with low lichen abundance Trees with low lichen abundance as well as trees that do not look like the ones in the photos can be assessed by counting the number of 5 g clumps that are present below 4 5 m Page 21 of the Field Guide shows photos of 5 g clumps and relates the number of clumps to Lichen Class How do we know how much lichen is really on the trees in the photographs The lichens on the trees in the photos were removed in stages air dried 25 to room temperature and weighed First some of the branches and Fstimate Estimate a lichens that were invisible to the camera were removed and the tree was e SAEN photographed Then a portion of the remaining lichens was removed l from each part of thetree bagged and labelled and
22. are only estimates Scoring may not always be done consistently and even if it is the Lichen Classes are very broad at the upper end of the scale H owever biomass estimates are valuable for developing and comparing timber harvesting prescriptions that attempt to reduce the impact on lichen abundancein thestand 25 9z table 4 Lichen biomass summary for subalpine fir B and spruce S based on Tables 2 and 3 Lichen Class 1 Lichen Class 2 Lichen Class 3 Lichen Class 4 Lichen Class 5 All Lichen Classes dbh B S Total Total B S Total Total B S Total Total B S Total Total B S Total Total Total Total class s ha s ha s ha gha s hat s ha s ha gha s ha s ha s ha g ha s hat s ha s ha g ha s ha s ha sha gha gha yha 10 49 5 0 0 49 5 123 8 70 1 8 2 78 3 1 957 5 20 9 0 0 20 9 3 135 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 216 3 148 7 15 5 1 55 106 265 30 4 5 5 35 9 897 5 60 8 0 0 60 8 9 120 0 5 1 0 0 5 1 2 295 0 0 0 0 0 0 0 0 0 12 339 0 112 4 20 0 0 0 0 0 0 0 0 10 3 17 7 28 0 700 0 61 6 8 8 70 4 10 560 0 10 3 0 0 10 3 4 635 0 0 0 0 0 0 0 0 0 15 895 0 108 7 25 0 0 0 0 0 0 0 0 5 9 53 11 2 280 0 26 6 10 6 37 2 5 580 0 14 8 0 0 14 8 6 660 0 3 0 0 0 3 0 2 400 0 14 920 0 66 2 30 0 0 0 0 0 0 0 0 0 0 0 9 0 9 22 5 37 4 1 8 39 2 5 880 0 9 4 1 8 11 2 5 040 0 0 0 0 0 0 0 0 0 10 942 5 51 3 35 0 0 0 0 0 0 0 0 55 0 0 5 5 137 5 13 8 6 1 19 9 2 985 0 8 3 12 2 20 5 9 225 0 2 0 0 0 2 0 1 600 0 13 947 5 47 9 40 0 0 0 0
23. cheme because it is logical to allow for the possibility of trees with no lichen at all and because there are some occasions when it is actually needed Amount of Lichen grams A tree that looks like a high Class 3 but has 1 3 as much additional lichen on branches that cannot be N PA seen from that viewpoint belongs in Class 4 lt N N N N N Nn amp 8 5 5 amp V U v UO s g fa m i l fa o o f fa o 3 3 3 3 3 5 50 100 200 250 300 400 500 625 175 T A tree that looks like a low Class 4 300 600 but is missing half the crown belongs A in Class 3 A tree that looks like a high Class 4 but is missing half the crown still belongs in Class 4 figure 2 How to use the scale to adjust lichen scores 4 5m 2 25 m The 4 5 m point can be estimated with reference to the height of a person 45 1 5m 4 5m 45m A clinometer can be used to identify the 4 5 m point There are more lichens on one side of the tree than the other Which side do I look at when I assess the tree It helps to understand that the Field Guide photographs were taken so that as many of the lichens as possible would be visible Some of the lichens that could not be seen were removed before the photos were taken Thus the lichens that make up the biomass values given on page 21 in the Field Guide are all or nearly all visible in the photos Usually it is adequate to position
24. chen identification Similar looking species may be lumped with Alectoria and Bryoria for the purposes of determining Lichen Classes Users who would liketo learn more about lichen identification may consult aField Guide 4 Rominger E M C T Robbins and M A Evans 1996 Winter foraging ecology of woodland caribou in northeastern Washington J Wildl Manage 60 4 719 728 5 McCune B and T Goward 1995 M acrolichens of the northern Rocky M ountains Mad River Press Inc Eureka Calif Vitt D H J E Marsh and R B Bovey 1988 M osses lichens and ferns of northwest North America Lone Pine Edmonton Alta 4 BASIC CHARACTERISTICS OF LICHEN DATASETS Thefrequency distribution of actual lichen biomasson treesin atypical ESSF stand is strongly asymmetrical there aremore trees with small amounts of lichen than with largeamounts When lichen abundance is rated using the Field Guide the data tend to be moresymmetrical resembling anormal bell shaped distribution H owever the appearance of anormal distribution is artificial It results from using Lichen Class scores with an equal interval between them 1 2 3 etc to represent different sized rangesin biomass 0 5 5 50 50 250 etc Thishasimportant consequences for data analysis Usually oncea dataset of any type has been obtained thenext step isto summarizeit with asinglevaluesuch asthe mean or the mode most prevalent score to represent the data asa whole T
25. ct paths planned to cover a mountainside survey area of about 1000 Na wane cssccccsssssssssssssessssssesssssssesssssssessessusseesssssssteessssssetsssssseetiesssee 15 9 Lichen abundance zones based on the distribution of lichen plot scores in the survey area shown in Figure8 10 Trees per hectare by Lichen Class based on Tables 2 and 3 11 Lichen plot scoresin a group selection block and harvest openings planned to maximize lichen retention after the first harvest entry o cccsssssssssusesssssussssessssssssessusssssessssssasssssssessecssssessesieseeses 28 12 Trees per hectare in Lichen Classes 1 5 before and after a single tree selection harvest u s 30 13 Effects of two selection harvesting prescriptions on post harvest lichen abundance and losses 32 vi 1 PURPOSE OF THE HANDBOOK Background Arboreal lichens lichens that grow on trees are critically important winter forage for caribou in parts of British Columbia Arboreal lichens grow abundantly on some old trees but are generally sparse on young trees Forest managers can make better decisions about whether where and how to log in caribou habitat if they haveinformation about the abundance of these lichens The Field Guide Estimating the Abundance of Arboreal Forage Lichens describes a method of estimating lichen abundance on individual trees The Field Guide presents a series of photos of trees with known amounts of lichen bel
26. dominant lichen genus and map discontinuities in genus composition if they exist Alectoria and Bryoria differ in their ecological tolerances and are likely to respond somewhat differently to treatments that open thecanopy 22 Todeterminethecontribution of dead treesto lichen abundance If dead trees are an important substrate for available lichen in the stand then a harvesting method should be selected that will permit retention of some of thosetrees without endangering workers e Torelatelichen abundanceto tree species and diameter A tablethat relates lichen abundance to tree species and diameter can be used to predict the effects of various possible selection harvesting prescrip tions on lichen abundancein the residual stand Sampling methods Lichen assessments at the block level are best donein conjunction with atimber cruise Both kinds of data needed for a block level analysis can then be obtained at the same time and compiled jointly Lichen scores tend to vary considerably from plot to plot and from tree to tree Gener ally 25 30 plots of 5 10 trees each have been found adequate to charac terize a homogeneous area regardless of size If a block is heterogeneous in lichen abundance it should be stratified into sampling units that are more homogeneous within than between and each stratum should then be sampled separately M ore detailed information about sampling methods is found in Section 5 Prism plots may be used p
27. efore deciding how to distribute the sample among them If a reconnaissance confirms that the site is highly variable stratifying the sample by tree types or classes may also help to reduce extraneous variation provided that good strata can be defined This is an important qualification since strata that are mismatched to actual sources of variation will yield less efficient estimates than a simple random sample From datasets that have been examined so far few good stratification criteria have emerged Some of the more consistent ones are tree species wildlife tree class tree size dbh class age class tree or stand Differences in such variables as tree characteristics within plots often account for much of the variation among plots If sampling strata cannot be identified beforehand post stratification by these factors may be feasible provided the resulting strata are not too small and the variation within them is less than the variation among them To compile the data for analysis the same basic steps described in Sections 4 and 5 apply H owever analyzing the data for research pur posesislikely to extend beyond comparingsites or parts of sites and theanalytical methodswill be more complex To control for themulti dimensional variation in lichen scores log linear modelling may bethe method of choice and for comparing lichen scores with external vari ables logistic regression will beappropriate See Yates F 1981 Sam
28. equency distribution of Lichen Classes at two ESSF sites that have low lichen abundance and arenot used as winter range by mountain caribou Only 7 and 9 of thetwo samples respectively have lichen scores of Class 3 or higher Although graphingthedataisoften enough for basic comparisons such as these the differ ences may not always be as distinct as between Figures 6 and 7 or thesamplesizes may be small raising the possibility that the result is dueto chance Then a statistical test may be required as discussed in Section 5 All the information needed for formal tests on score distributions is contained in the frequency table H owever the user may also wish to assess how scores are distributed over different parts of thesite or how much lichen is present per unit area For these purposes 7 When samplesizes differ as at these sites the score counts should be plotted as percentages When compared by statistical methods described in Section 5 the counts themselves are used unequal sample sizes are automatically taken into account 11 further information is required To examinespatial variation it isnecessary to record thelocation of each score To estimate total amounts of lichen conversion factors arerequired Section 5 to express thescoresas biomass equivalents Biomass equivalents are especially useful for planning and evaluating harvesting prescriptions at theblock level 5 USING THE FIELD GUIDE FOR RECONNAISSANC
29. eresults of a single tree selection prescription that could be used in the pre harvest stand described in Tables 2 and 3 In thissampleprescription each diameter class between 17 5 and 52 5 cm in the post harvest stand has 1 3 times as many stems asthe next larger diameter class stemslarger than 52 5 cm dbh areretained All dead trees are removed Of theliving trees the Lichen Class distribution of merchantable trees gt 17 5 cm isapproximately maintained Figure 12 In Lichen Class 3 or greater 332 areretained 75 of the 446 stens hain the pre harvest stand Overall the analysis predicts that implementing this prescription would result in 82 kg haof arboreal lichen remaining Table 7 compared to 122 kg hain the pre harvest stand Table5 In thisexample only about 9 of theavailablelichen in the pre harvest stand occurred on dead trees Table 5 In somestands dead treesmakea much larger contribution to thelichen forageresource W here dead trees arean important source of availablelichen managers should consider a prescription that will allow someto beretained Group selection may offer opportunities to retain dead trees in the leave areas where they cannot fall into work areas In someblocksin caribou habitat the Workers Compen sation Board has granted variances to selectively retain dead trees during harvesting Any prescription involving retention of dead trees should be developed in consultation with the Workers Compensation
30. esto lichen abundance 400 350 Legend 300 Subalpine fir 250 a HE Spruce 200 Snags v Trees per hectare fo 1 2 3 4 5 Lichen Class figure 10 Trees per hectare by Lichen Class based on Tables 2 and 3 27 Legend Number of trees plot of Lichen Class gt 3 low 0 1 re medium 2 3 Ww higmhes g harvest openings OJo Oo an O zm Z figure 11 Lichen plot scores in a group selection block and harvest openings planned to maximize lichen retention after the first harvest entry Prescription planningin mountain caribou habitat is discussed in M oun tain Caribou in M anaged Forests Preliminary Recommendations for M anagers Where basal area selection harvesting is being planned stand tables can be used to predict the effects of various possible prescriptions A spreadsheet with linear programming capability is ideal for this Enter the Lichen Class data broken down by treespecies and diameter class asin Table 4 Then test various scenarios by varyingthenumber of stems hain each species diameter class as called for in a potential prescription In most stands a prescription that is suitable for caribou habitat should aim at least to preserve the same relative frequency distribution
31. forage Arboreal lichens form part of the winter diet of caribou in many parts of British Columbia The caribou of southeastern and east central British Columbia known as mountain caribou are especially dependent on arboreal lichens For caribou living elsewhere in British Columbia often called northern caribou arboreal lichens are one of several important winter foods The two types of caribou belong to the same species but occupy different winter habitats Figure 1 With their well insulated coats and large hooves caribou are well adapted to cold weather and deep snow Mountain caribou have an unusual strategy for surviving the long snowy winters Unlike moose and deer which seek out shallow snow mountain caribou spend much of thewinter in high elevation forests Deep snowpacksat high elevations bury other forage that caribou might use but also createa platform that allowsthecaribou to reach higher into thetrees This strategy of spending winter at high elevations has the advantage of separating the caribou from moose deer and wolves most of the time but it does make them depend ent on asingletype of forage arboreal lichens Mountain caribou do not necessarily spend the entire winter in the high elevation ESSF zone In some areas they also use the lower elevation Interior Cedar H emlock ICH zone especially during early winter when the snow is fresh and soft In the ICH zone caribou usually cannot reach many lichens on the lo
32. hens from the lower branches of trees in other biogeoclimatic zones as well The Field Guide may be used in any zone where it is important to assess lichen abundance on the lower 4 5 m of trees H owever users should be aware that e Only thelichens within about 1 5 m of thetop of a settled snowpack are available to caribou e Treesin other forest types may not look like the trees in the photos For example lodgepole pine in the M ontane Spruce M S zone may have few branches below 4 5 m but support a dense lichen cover on the trunks Users should not assume that assessments based on the Field Guide are good indexes of lichen abundance on entire trees Lichen abundance in the lower 4 5 m of atreeis not necessarily well correlated with lichen abundance in the rest of the crown THIS HANDBOOK IS INTENDED TO HELP MANAGERS e plan and carry out data collection using the lichen abundance Field Guide e use the data as a tool for planning in caribou habitat In the ESSF zone the deep snowpack allows caribou to reach lichens on the lower branches of trees The Field Guide was developed for use in the ESSF but it may be used in any zone where it is important to assess lichen abundance on the lower 4 5 m of trees In the MS zone more lichens may be available to caribou on the trunk of lodgepole pine than on the branches 2 WHY ASSESS LICHEN ABUNDANCE Caribou depend on lichens for winter
33. hisis not recommended for lichen scores particularly if theaim isto estimate lichen biomass T here aretwo reasons for this Oneisthenon linear relationship between scores and biomass scores of 2 and 4 differ equally from a mean score of 3 but they do not correspond to equal differences in biomass Thesecond reason has to do with the way caribou forage If trees scoring 3 or more are chosen for feeding but treesscoring 1 or 2 areignored themean score for all trees com bined would bea poor measure of their relative valueto caribou A better measure would bethenumber of trees with scores of 3 or more Such information islost when the data are merged or simplified into a composite value such asthemean Therefore when a set of lichen scoresis summarized an important piece of informa tion to retain istherelative abundance of different scores Thiscan bedoneby compiling thedata as a frequency distribution tableof the 6 The relationship between scores and biomass is described in Section 6 120 Co an Number of trees D N 0 100 200 300 400 500 600 700 800 900 1000 Lichen biomass g figure4 Frequency distribution of Alectoriaand Bryoria biomass in the lower canopy of trees in an ESSF stand 70 60 50 40 30 i m 0 1 2 3 4 5 Lichen Class Number of trees figure 5 Frequency
34. ildlife trails may also be misleading because the trails seldom run through typical parts of the site Unless the area is rather small the minimum 150 trees may be ob tained long before the area has been traversed Transects through a large area should therefore be sampled at intervals rather than scoring every tree encountered on the line One way to do this would beto determine the total length of the path to be traversed select 30 distances along the transect at random and place a plot at each predetermined location as the transect is walked Figure 8 Local samplingor smaller sites When time or distance to be traversed is not the primary concern a two dimensional grid may be more suitable A grid layout is more likely to spread the sample over all portions of the site and is easier to stratify if needed M ap the site with grid lines 10 m apart and placea plot at each of 30 randomly chosen intersections 30 per stratum if the site is being split into sub units Usefixed area plots not prism plots If prism plots are used the Lichen score frequencies obtained from fixed area plots require no data must be corrected for adjustment for tree size dbh However prism plots select more large differences in the probability trees than small trees and large trees often have more lichen Prism plots of sampling trees of different should not be used for sampling lichen unless they are part of a cruise sizes that will be corrected for bias
35. in tree selection 9 Stratification is discussed further in Section 8 14 Boundary of survey area Observer 1 nee drop off ee 4970 4880 4500 1a pomi pickup oint P 4900 5010 870 Observer 2 drop off point Legend b lt t Transect path Observer 2 pickup 5240 5020 1169e Lichen plot and pom 5090 4940 distance m 500 m figure 8 Lichen plots at random distances along two transect paths planned to cover a mountainside survey area of about 1000 ha 15 Settheminimum diameter of sampletrees at 7 5 cm dbh Usually theminimum diameter limits used in forestry surveys arealso suitable for lichen surveys H owever some types of vegetation survey may include very small trees in the sample Since very small trees usually have low lichen abundance including them can resultin inflated frequenciesin the lower Lichen Classes A minimum diameter of at least 7 5 cm is sug gested for sample treesin lichen surveys Compiling the data A spreadsheet with a count frequency accumulator and graphing capabilities will be sufficient to handle all of the tasks described below Determinethefrequency distribution of lichen scores For each area or stratum sampled determine the total number of stems in each Lichen Class and produce a frequency table Table 1 Table the frequencies according to any categories that will beimportant in your analysis such as tree species or
36. itish Columbia and the H abitat Silviculture Protection Fund adminis tered by the British Columbia M inistry of Environment Lands and Parks Omineca Peace Region We thank Dave King Ministry of Environment Lands and Parks for his support of the project Ron Osland Cariboo Forest Region for his advice on timber cruising procedures and Vera Sit Ministry of Forests Victoria for her com ments on statistics Michaela Waterhouse Cariboo Forest Region Wes Bieber Weyerhaeuser Canada Ltd Brian Harding Northwood Pulp and Timber Ltd and Jocelyn Campbell University of Northern British Columbia allowed us to use their data Dave King Mark Phinney and M ichaela Waterhouse reviewed earlier versions of the manuscript and the proofs were edited by Fran Aitkens Graphics and layout were produced by The Borealis Communications Group CONTENTS How Users can Contribute to Adaptive M anagement ssesssssssssssessesessssssuesssusssssussseussstnsssssenesssenesten iil Acknowledgements ovvccccccssssssssssssssssssssesssssssessessssessssssssstsessssssstssssssseeissssssesenssoseeaessseeeesssastetesssssssneisessseetissssseeteasset iv 1 Using the Field Guide for Research and Inventory Purpose of the Handbook 0 0 ccessscessssesssssnssssssessesssssssessssssusssesssssseausessusnssssnsateessstussessesessesussesssnseteaiate 1 Background o ecssssssssesstsessssseneesstnssetsssesesee Who isthe handbook for u s Whierecaniit DEUS
37. lable at different times of the year During winter caribou tend to range over large areas moving frequently and unpredictably This behaviour makes it difficult for predatorsto find them Theamount of spacethey need to avoid predatorsis probably greater than theamount they need just for foraging However to serve as suitable habitat for avoiding predators an area must also provide accessible food Accessibility varies with depth and density of snow To maintain caribou populations managers must ensure that large areas with accessible forage are available in the right places at the right times Information about lichens may be required for approval of plans Increasingly the B C Ministry of Environment Lands and Parks is requiring that caribou habitat be considered in forest management plans Information on lichen abundanceis an important input to manage ment decisions in a planning area that includes caribou range In some areas information about lichens may be requested before a man agement plan is submitted or may increase the likelihood that the plan will be approved In several regions Ministry of Environment Lands and Parks guide lines call for use of selection silvicultural systems to maintain caribou habitat in certain areas Some land use plans now specify partial cutting or modified harvesting instead of clearcutting in caribou range In such cases information on lichen abundance may be essential to the design
38. ld M Armleder R P F R P Bio Cariboo Forest Region Suite 200 640 Borland St Williams Lake BC v2g 4t1 for B C Ministry of Forests Cariboo Forest Region Suite 200 640 Borland St Williams Lake BC v2g 4t1 Published by B C Ministry of Forests Forestry Division Services Branch Production Resources 595 Pandora Avenue Victoria BC v8w 3e7 1998 Province of British Columbia Copies of this and other Ministry of Forests titles are available from Crown Publications Inc 521 Fort Street Victoria BC v8w 1e7 HOW USERS CAN CONTRIBUTE TO ADAPTIVE MANAGEMENT The recommendations in this handbook are based on studies of arboreal lichen datasets collected from various sites in three regions of east central British Columbia Although these data proved to be fairly diverse they represent only a small portion of the variety that occurs in mountain caribou ranges Data from additional sites would help to fill in the picture of natural variation in arboreal lichen abundance This would improve future recommendations for sampling and for manage ment Users of the lichen Field Guide are encouraged to assist this process by sharing the data they obtain using the methods recom mended here A form for submitting this information is found at the back of the handbook ACKNOWLEDGEMENTS Preparation of this handbook was funded by the British Columbia Ministry of Forests Cariboo Forest Region Forest Renewal Plan of Br
39. lichen genus composition Show thesamplesize Include the plot size and the number of plots or transects per stratum Table 1 The plot size is needed if lichen scores are to be expressed per unit area or compared with stem densities Stem density can be calculated for each Lichen Class per stratum by dividing the total number of scores in each class by the combined area of the sample plots in that stratum A compilation that includes the frequency distribution and the sample size number of plots and plot size retains all the basic information in thesample table 1 Frequency distribution of Lichen Class scores with associated information A gt 70 Alectoria AB mixed Alectoria and Bryoria B gt 70 Bryoria Genus composition is not recorded for Lichen Class 1 trees Lichen Class 0 1 2 3 4 5 Tree species Subalpinefir 0 17 37 76 27 2 Spruce 0 3 B 20 9 0 Snags 0 5 5 14 2 0 Total no of scores 0 25 55 110 38 2 Genus composition A 16 37 15 0 AB 35 71 23 2 B 4 2 0 0 Total no of scores 55 110 38 2 Stems ha 0 83 183 367 127 7 Total plots 30 Plot size ha 0 01 16 Frequency distributions should be analyzed by methods designed for proportions or counts Analyzing the data Lichen abundance can be assessed for a site or compared amongsites or compared amongstrata by e thedifferencesin therelative frequencies of different scores e thedifferences in the percentage of sc
40. ly Determining the sample size Oncethe sampling domain has been defined the next step is to deter mine the number of trees to be scored within it This is set by the desired level of assurance that the sample scores will reflect the scores on all the trees present Theminimum samplerecommended is 150 trees per domain This will provide 95 assurance that the sample frequency for each score will differ by no morethan 10 from itstruefrequency in thedomain asawhole If plots are used aim for fiveto 10 treesin each plot Fewer than five trees istoo few to encompass the typical variation in astand and morethan 10 trees seldom adds much further variation Circular plots of 0 01 ha radius 5 64 m will usually be adequate for stands with about 600 stems ha or more In stands with 300 600 stens ha 0 015 ha plots radius 9 91 m increase the likelihood of obtaining the minimum five trees At an average of five trees per plot about 30 plots will be required to obtain 150 trees in total Evidence from existing datasets has shown that the variation from plot to plot will usually stabilize once 25 30 plots have been sampled as long as the area as a wholeis relatively uniform i e no obvious patchiness in lichen abundance or occurrence If abundance varies widely from place to place the area should be sub divided into more homogeneous portions and 30 plots taken from each portion If a sample of fewer than 150 trees has already been scored its
41. management trialsisan important part of thelearning process Contributing to the knowledge base on lichen abundance The recommendations in this manual are based on analysis of only a few datasets A larger database would serve two important functions e It would provide the basis for more complete recommendations about how to use the lichen abundance Field Guide as a tool e t would improve knowledge of the abundance and distribution patterns of arboreal lichens in caribou habitat Monitoring the effects of selection harvesting on lichen abundance The Field Guide can be used as a monitoring tool as well as a planning tool Assessments based on the Field Guide can be expected to detect only relatively large changes in lichen abundance Changes that result from damageto trees during logging or wind scouring of trees after logging should be detectable Changes that result from faster or slower growth rates after logging would probably not be detectable at least not for along time To monitor lichen abundance in operational situations carry out pre harvest lichen assessments as usual Permanently mark the plot or transect locations and the assessed trees so they can be relocated If possible mark and assess treesin an adjacent control area that will not beharvested Wait at least a year after harvest to re assess the trees Data resulting from lichen reassessment may be used to determine the effects of the harvesting on abundancein each
42. mples should not beignored Mapping the data Lichen score frequencies often vary between different parts of a site When this is suspected the data should be mapped and examined for signs of clustering A multiple frequency distribution cannot feasibly be mapped and so some form of data reduction or summarization is needed The percentage of trees per plot with Lichen Class scores gt 3 or the number per ha or km2 are mappable attributes A threshold Lichen Class of 3 is suggested because of evidencethat caribou aremorelikely to perceivetrees of Lichen Class 3 or greater as afood source Section 3 Sincethese counts or percentages focus on the upper end of theabundance scale they aremore biologically meaningful than averages based on the wholescale i e including low scoring trees that caribou ignore Figure9 shows how mapped lichen scores could be used to stratify a heterogeneous survey areainto two lichen abundancezones Using the data to manage caribou habitat At present not enough information is available on patterns of lichen abundancein caribou range to specify how much lichen must be present for a stand to be useful as winter foraging habitat Current research will help to address this question In addition users who collect lichen abundance data operationally in areas of known caribou habitat can help by registering their projects in a central data registry and contributing data to the information base if asked to do so A da
43. of a Sa In the ICH zone more lichens may be available on suitable Silviculture Prescriptions fallen trees or branches than are within reach on standing trees E Northern Caribou E Mountain Caribou Prince Rupert E YO 0 R 0 D figurei Range of caribou in British Columbia Vancouver 3 USING THE HELD GUIDE TO ASSESS INDIVIDUAL TREES Basic procedures The Field Guide is easy to use but it is important that it be used in the same way from day to day and by all crew members e Practice together to develop consistency At the beginning of a project all users should practice together until their results are converging As there will always be trees at the borderline between Lichen Classes there will always be some discrepancies between the scores given by different observers If training and practice have been adequate it is reasonable to expect that most observers will give the same tree the same score and that scores that do differ will not differ by more than one Lichen Class e Know the sampling scheme and follow it rigorously To avoid bias users must score all the sample trees and only thosetrees Dying and dead sample trees are also assessed Unusual trees that occur outside the plot or transect may be described in thefield notes but must not beadded to thedata Walk around the tree to find the best viewing position The lichen estimate applies to the entire lower porti
44. on of the tree not just to one side Lichen abundance may look different from different angles The photographs in the Field Guide were taken from the angle that best revealed the lichens The best viewing position is the one from which you can see the lichens best with the least interference from other trees If you cannot see all the lichens from one viewing position you must adjust your estimate to account for the additional lichens viewed from a different position e Includeall lichens below 4 5 m in the assessment This includes lichens on thetrunk of thetree and on branches that originate above 4 5 m but hang down into the zoneof lichen estimation On sloping ground the 4 5 m line runs parallel to the slope Caribou can reach higher into the canopy on the uphill side of the tree than on the downhill side of the tree In the Field Guide photos 4 5 m isindicated by the red mark at the top of theheight pole e Keep referring to the Field Guide To use the Field Guide first decide Remember that the photo which photo series A B or C best matches the tree Use it to guide shows trees that are on determine roughly into which class the tree fits Then confirm your the boundary between lichen assessment by turning to the photos that distinguish the classes classes Remember that all the photos in the Field Guide show trees at the boundary between Lichen Classes If you classify a tree as Class 4 for example it should have more
45. on risk The strategies of spacing themselves out over large areas using different winter ranges in different years and moving extensively within ranges in a single winter probably help caribou avoid predation Caribou may avoid using ranges where the risk of predation is high e Habitat fragmentation Parts of some ranges may not be used because they have become isolated from other parts or because the amount of suitable habitat has shrunk Forest management in caribou habitat should be planned so that large areas arein a suitable condition for use by caribou and are free from roadsand if possible snowmobile access Abundant arboreal lichens do notin themselves makean area suitable habitat but without them an area cannot beused as winter range by mountain caribou 6 USING THE HELD GUIDE FOR PLANNING AT THE BLOCK LEVEL Block level objectives for lichen assessment Usually the purpose of lichen assessment at the block level isto provide the information needed to select a silvicultural system and to plan a prescription Some of the objectives might be e To quantify overall lichen biomass in the block Overall lichen abundanceis an indicator of the importance of the stand as potential winter range for caribou e To identify and map discontinuities in lichen abundance Information on the spatial patterning of lichen abundance and of genus composition can be used to plan harvest block locations and boundaries e To identify the
46. ores above a certain limit e the cumulative score total s or biomass total s for the site s or categories e thetotal or average biomass per plot These are not the only possibilities H owever the first step in any of these analyses should be to produce bar graphs showing the number of scores per Lichen Class for each category of interest Bar graphs quickly indicate the main features of each sample such as the most common score how typical it isfor the sample as a whole and how it compares to the most common scorein other samples If these features are distinct and the sample sizes are adequate a graph will often be enough to answer the basic question of interest for example whether two sites differ in lichen abundance But if the samples are small or the difference between them does not appear distinct amore formal statistical test may be required The common purpose of the statistical tests described below is to gauge the likelihood that the results observed could have occurred by chance instead of representing a real biological effect The calculations required are not complex A spreadsheet is sufficient for handling them and procedures can be found in most statistical texts Comparing relativedifferencesin lichen abundance To find out whether the number of scores differs significantly among Lichen Classes their relative frequencies can be compared This can be donefor a single sample comparing class with class or for t
47. ose It has been widely adopted but users have experienced some frustration in trying to apply the assessments described in it Although there are many unknowns still to be overcome this handbook isan important step in making the Field Guide a more useful tool for assessing lichen abundance 14 Vegetation Inventory Working Group 1997 Vegetation resourcesinventory ground sampling procedures B C Ministry of Forests Resources Inventory Branch Victoria B C Draft M arch 31 1997 34 DATA REGISTRATION FORM Organization Address Contact person Telephone Fax E mail Name and location of project area NTS UTM Biogeoclimatic subzone s Purpose of project reconnaissance caribou habitat cutblock planning caribou habitat cutblock monitoring research Specify inventory Specify other Specify Area sampled eg three 40 to 60 ha blocks 5 000 ha watershed Brief description of sampling scheme e g prism plots systematically located in each of two strata Total number of plots or transects Year s and season s of sampling Assessor s previous experience with lichen Field Guide no previous lichen surveys 1 4 previous lichen surveys 5 or more previous lichen surveys Did you collect data on the types of lichens present yes no Data format
48. ow 4 5 m which is the part of the tree within reach of caribou in winter The user compares the tree being assessed to the photos and scores it as belonging in Lichen Class 0 5 The Field Guide is quick and relatively simple to use can give similar results when used by different people and produces assessments that are related to actual lichen biomass The Field Guide however does not offer direction on how lichen assessments fit into planning and data collection processes It does not discuss how to sample or what to do with the data collected This hand book is intended to fill those gaps and help managers use the Field Guide as a tool for planningin caribou habitat The Field Guide also has research and inventory applications and can be used to monitor the effects of forestry practices on lichen abundance This handbook is based on analysis of several datasets collected during thefirst two years after the Field Guide was published These include data from one reconnaissance level project in which several sampling methods were compared and several block level datasets collected for research or operational forestry purposes The data were collected in the Engelmann Spruce Subalpine Fir ESSF biogeoclimatic zone in three M inistry of Forests Regions As more data become avail able from more places the characteristics of lichen data can be described more fully The recommendations in this handbook may then be ex tended and refined
49. pling methods for censuses and surveys Griffin London U K or Cochran W G 1977 Sampling Techniques Wiley New York N Y 33 Inventory The Field Guidemethods have been incorporated into the vegetation inventory procedures of the Resources Inventory Committee If lichens are routinely included in surveys where caribou are known or suspected to occur alarge body of data may eventually result This will increase knowledgeof lichen abundance and distribution in morecaribou ranges than at present 9 CONCLUSION Since the late 1980s forest managers and researchers have been experi menting with timber harvesting methods that will better maintain arboreal lichens in mountain caribou habitat Developing forestry practices that will allow timber harvesting and also maintain habitat for caribou is a large scale and long term undertaking To a great extent its success will depend on the willingness of managers to apply the princi ples of adaptive management These include recognizing the present lack of knowledge about the effectiveness of various forestry practices in maintaining caribou habitat deliberately designing experiments to test new practices and using the results to modify management practices thereafter Using lichen abundance data for planning and monitoring purposes in caribou habitat areas is one way that managers can contribute to the process The Field Guide for assessing lichen abundance was published for that purp
50. presents one tree stem density is simply the number of scores recorded per plot divided by plot size If plot size isthe same at each site it can be ignored and the comparison can be based directly on the number of scores plot by plot Two tests suitable for comparing stem densities between different samples are e Kruskal Wallis AN OVA Analysis of Variance for two or more independent samples e g samples from different sites e Friedman ANOVA for related samples e g samples from the same site before and after a management treatment The Kruskal Wallis test and the Friedman test have a similar rationale and are performed on theranks of thelichen estimates rather than the estimates themselves Comparingabsolute differences in lichen abundance There may be situations in which users wish to calculate lichen abun dance on an absolute basis Lichen abundance expressed as biomass per unit area integrates the effects of lichen abundance and stem density into a single measure Expressions of biomass per unit area make it easy to compare overall lichen abundance between sites but not to determine whether differences are due to lichen scores stem density or both To compare lichen abundance per unit area between sites convert the score value for each tree to its biomass equivalent using the conversion factors given in Section 6 Using these values calculate the total lichen biomass for each plot The outcome may look like a dataset with
51. r two but chi square and G tests can easily be extended to threeor more For the chi square test however the limitation at small scores still applies Thus for comparing three or more frequency distributions simultaneously the recommended method is thelog likelihood ratio G Differencesin stem density The tests described above compare only the relative numbers of scores in each Lichen Class Thus aG or K S test will show no difference between sites with the same ratios of scores even if one of the sites has a much larger total number of scores than the other Sokal R R and Rohlf J F 1995 Biometry Freeman amp Company New York N Y The factor q is defined in this reference 18 Lichen scores expressed as biomass do not meet the assumptions of parametric statistics and should be analyzed using non parametric methods Sites with similar score ratios might differ in score totalsfor either of the following reasons e moreplots were sampled at onesitethan the other or e thetwo sites differed in stem density In comparing score ratios the number of plots is immaterial except to determine the sample size for a G or K S test However a difference in stem density would mean a difference in the amount of lichen per unit area even if each site has the same amount of lichen per tree Thus when samples differ in total number of scores they should be examined for possible differences in stem density Since each score re
52. reliability can be assessed by calculating the 95 confidence limits CL of each score frequency If the lower CL for any frequency is zero or the upper CL is more than 10 greater than the frequency itself additional trees should be scored or anew and larger sample taken 13 Laying out the sample To provide unbiased results any treesampled must havethesamechance of being chosen as any other even though the 150 trees may be just a small minority of thetotal trees present Thesample must therefore span thewhole of thesurvey area and thesampling points should berandomly placed regardless of whether alinetransect or atwo dimensional array of plots is used Transects for reconnaissance purposes Transects are an effective way to conduct the first reconnaissance of a site No time is lost in setting up an array of plots and data are collected continuously while the area is being traversed Findings from the initial survey can show whether a more intensive or more local survey would be worthwhile and whether the area should be divided into strata to be sampled separately Transects are also well suited for sampling lichens along an environ mental gradient such as elevation or moisture regime Except to sample a gradient however the user must ensure that thetransect route does not follow some geographic feature that might affect the amount of lichen present Roads or stand edges are obvious examples but samples ob tained along w
53. rescription windthrown landings harvest trees Prescription 1 E Prescription 2 figure 13 Effects of two selection harvesting prescriptions on post harvest lichen abundance and losses 8 USING THE FIELD GUIDE FOR RESEARCH AND INVENTORY Research The Field Guide has a variety of uses in research such as e studying thecharacteristics of habitats used by caribou e studying therelationship between lichen abundanceand environ mental variables e studying the effects of forestry practices on caribou habitat attributes Sampling strategy for research purposes is likely to differ from sam pling for forest management planning or operations For planning or Operational purposes the sampling plan is mainly designed to quantify lichen abundance per se and other factors if sampled at all are sampled contingently For a research study the sampling design will depend on the type of question being asked the scale at which it applies and the sampling requirements for variables other than lichen abundance Since these considerations are bound to differ from case to case comments here about sampling design are limited to a few basic generalities Lichen abundance data tend to be highly variable and the sample characteristics tend to differ from site to site Regardless of whether lichens are the main or the contingent variable the main requirement of the sampling plan will beto gain as much control as possible over the 32 sources of
54. rovided the lichen data are compiled so as to correct for differences in the probability of sampling trees of different sizes as described below If the assessments are not being done together with a timber cruise then fixed area plots are recommended As at the reconnaissance level 0 01 ha radius 5 64 m plots are recommended for stands with morethan 600 stems ha and 0 015 ha radius 9 91 m plots for stands with 300 600 stems ha The cruise tally sheet does not provide a column for Lichen Classes The Ministry of Forests recommends that the scores be recorded in Column 60 However because the numbers 1 7 have already been assigned to another usein that column alphabetic codes should be substituted for Lichen Class numeric codesas follows Lichen Class Enter in Column 60 0 A 1 B 2 C 3 D 4 E 5 F Thissubstitution will ensurethat thereisno confusion with other codes which could affect appraisal values Theuseof Lichen Class codesin Column 60 should benoted under Remarks 23 Compiling the data WARNING Other codes used Creating stand tables in Column 60 of the cruise The easiest way to compile lichen data for a stand level analysisis tally sheet affect the through a special run of the provincial cruise compilation program At appraisal value of the stand present the provincial program hasnot been modified to accommodate Users must ensure that lichen data H owever the changes needed are minor and someforestry compila
55. ta registration form can be found at the back of this handbook Where site conditions slope topography and stand structure are suitable for caribou winter range lichen abundance data can be used to rank stands for importance as potential foraging areas for caribou This information can be used to guide decisions about whether a particular stand should be harvested and what silvicultural systems should be considered 20 Boundary of survey area Lower lichen re TA abundance o i O Higher 1 5 lichen abundance Legend Number of trees plot of Lichen Class gt 3 low 0 1 re medium 2 3 high gt 3 500 m in Figure 8 figure 9 Lichen abundance zones based on the distribution of lichen plot scores in the survey area shown 21 Lichen abundanceis only one factor that affects how caribou usean area in winter and itis only one factor to consider in planning forest manage ment Some of the other factorsthat affect caribou use are e Access Road access into caribou ranges can make caribou more vulnerableto legal and illegal hunting and can increasethelikelihood of disturbance Ploughed roads or packed trails can increase access by wolves at times when the snowpack is soft e Disturbance Caribou have abandoned some winter ranges that became heavily used by snowmobilers Heli skiing and cross country skiing may also be significant disturbance factors e Predati
56. than to imagine lichens that are not there To some extent observers can compensate for this tendency by being aware of it How much lichen must a tree have for caribou to use it Research into the foraging behaviour of caribou indicates that caribou are more likely to feed from trees with abundant lichens than trees on which lichen issparse The Field Guidewas not used to assess lichen abundanceon thetreesin thosestudies Based on their experience Terry E 1994 Winter habitat selection and foraging patterns of mountain caribou M Scthesis University of British Columbia Vancouver B C Rominger E M C T Robbins and M A Evans 1996 Winter foraging ecology of woodland caribou in northeastern Washington J Wildl M anage 60 4 719 728 however theresearchers judged that treesin Lichen Classes 0 1 and 2in the Field Guidewerelikely to be passed by whereas trees in Lichen Class 3 or higher werepotential feeding sites Should I collect information on the types of lichens that are present and if so how There areimportant differences between the two types of forage lichen Alectoria and Bryoria that can affect forest management in caribou habitat First the two types seem to prefer different combinations of light and moisture and Bryoria is likely to respond more positively than Alectoria when stands are opened up by partial cutting Second thereis growing evidence that caribou choose Bryoria rather than Alectoria
57. the tree was photo graphed again This process was repeated until a series of photographs had been taken and all the lichens had been removed The lichens in the bags were weighed and theinformation used to calculate the biomass of the lichens present on the treein each photo N T 59 wn T i T Percentage of trees How accurately can observers classify lichen abundance on trees This question is being answered as part of an ongoing study of arboreal lichen ecology near Prince George An experienced observer used the 4 Field Guide to assess lichen abundance on 27 Engelmann spruce and oL ma subalpine fir trees Lichen abundance in the assessment zone of each tree 2 1 0 1 2 was then determined through detailed sampling and weighing Figure 3 Lichen Class estimated minus Lichen Class measured shows how the Lichen Class assigned by the observer compared to the 7 figure 3 Frequency distribution of the Lichen Class based on biomass sampling difference between Lichen Class Although most trees were classified correctly the errors that did based on measured samples and occur were nearly all underestimates rather than overestimates This Lichen Class assessed by an pattern is consistent with the results of other studies in which estimates observer of lichen biomass have been related to actual measurements It probably occurs because people are more likely to miss seeing lichens that are present
58. tions that include consultants havealready madethem They generatea stand tablethat lichen codes do not shows the number of stems per hectarein each Lichen Class by diameter improperly affect appraisal class Separate tables may be produced for different tree species and values different timber types or for other strata within theblock Thetables should be compiled in a general format such as ASCII text so they can be easily exported to a spreadsheet and re formatted as required for further analysis Tables 2 and 3 are examples of stand tables by Lichen Class for a proposed block with two tree species table 2 Stand table for subalpine fir by Lichen Class Subalpinefir trees per hectare dbh class Lichen Class 1 Lichen Class 2 Lichen Class 3 Lichen Class 4 Lichen Class 5 Total 10 49 5 70 1 20 9 140 5 15 5 1 30 4 60 8 5 1 101 4 20 10 3 61 6 10 3 82 2 25 5 9 26 6 14 8 3 0 50 3 30 37 4 9 4 46 8 35 5 5 13 8 8 3 2 0 29 6 40 10 0 20 0 30 0 45 16 3 8 1 24 4 50 2 5 10 1 12 6 55 1 5 3 0 3 0 7 5 60 1 1 1 1 65 2 0 2 0 70 0 0 Live trees ha 57 1 123 7 253 5 89 1 5 0 528 4 Snags ha 8 2 16 4 46 5 8 2 79 3 24 table 3 Stand table for spruce by Lichen Class Sprucetrees per hectare dbh class Lichen Class 1 Lichen Class 2 Lichen Class 3 Lichen Class 4 Lichen Class 5 Total 10 8 2
59. transectsin which every tree on thetransect issampled Often bar graphs are all that is needed to tell whether sites are similar or different in lichen abundance 12 ioe From theinformation availableto date itisnot possibleto say that any of these methodsis better than theothersfor surveying lichen abundance In choosingamongthem themajor considerationswill usually bethe objectives of the survey the size and shape of the area and the compatibil ity with methods used to gather other data at thesametime eg timber cruising Commonly lichen assessments will befitted into asampling protocol already developed for another purpose Defining the sampling domain Whatever method is chosen the first step in a sampling plan is to define the sampling domain also termed the sampling frame or sampling stratum Thisis the population of trees from which the sample will be taken M ost often the domain will be all trees gt 7 5 cm diameter at breast height dbh in some defined area such as a portion of caribou range or a particular forest stand or a proposed cutblock within it H owever the area might consist of distinctly different parts in which lichens might be more abundant in some than in others or the aim of the survey might beto compare different tree species sizes or ages within the same area Each such category is a separate domain requiring an adequate sample size even if the samples for each are collected simultaneous
60. wer branches of trees and instead eat lichens that have fallen onto the snow or are available on fallen branches or fallen trees They also eat the leaves of low evergreen shrubs such as falsebox or bunchberry Unlike moose and deer caribou browse very little on woody parts of shrubs Assessment of winter forage for caribou in theICH zone should include litterfall lichens lichens on fallen trees and vascular forage as well as lichens in the lower canopy Elsewhere in British Columbia where the snowpack is not as deep northern caribou can scrape away snow to get at other types of forage during winter in particular terrestrial ground lichens grasses and sedges N orthern caribou use arboreal lichens too but not as regularly as do mountain caribou Terrestrial vegetation may become inaccessible when the snow is deep or crusted and northern caribou then moveto forest types where they can feed mainly on arboreal lichens Sometimes northern caribou switch back and forth between terrestrial vegetation and arboreal lichens in a single forest type People sometimes wonder whether it makes any difference to caribou if the lichen supply is reduced by logging since arboreal lichens seem to be abundant and widespread in caribou range M ost studies of caribou in British Columbia have concluded that caribou numbers are not directly limited by the amount of food available to them What seems to be more important is the amount of suitable habitat avai
61. when both are available Information collected on the types of lichens present will contribute to a knowledge base that can be used to improve manage ment For most purposes it is adequate to classify the sample trees into one of three categories such as e A mainly Alectoria morethan 70 Alectoria e AB mixed Alectoria and Bryoria e B mainly Bryoria morethan 70 Bryoria The photographs on page 3 of the Field Guide will help If more detailed assessments are needed genus composition may be estimated to the nearest 10 Genus composition is usually recorded only for trees of Lichen Class 2 or more How can tell Alectoria and Bryoria from other lichens In the ESSF zone Alectoria and Bryoria are the dominant beard like lichens growing on trees In other zones you are more likely to encoun ter lichens that could be confused with Alectoria and Bryoria Bryoria species are brownish greyish or almost black with branches that are round in cross section not flattened Several reddish brown species have recently been renamed as N odobryoria they should be included with Bryoria in lichen assessments Alectoria sarmentosa is yellow green to grey green Some U snea species look similar but havea tough central cord that is noticeable when a fragment is gently pulled apart Ramalina thrausta looks similar but the branches havehook shaped tips M ost often users of the Field Guide will not need to carry out de tailed li
62. wo or more samples comparing all classes jointly If the data are compiled as raw frequencies with no further conversion asin Table 1 they should be regarded as samples from a multinomial distribution and analyzed using methods for dealing with proportions or counts Contingency methods chi square or log likelihood ratio G are traditional ways for doing this 10 Various pre programmed routines e g SYSTAT SPSS and SAS further reduce thetask by generating probability levels internally thus avoiding theneed to refer to probability tables 17 m For assessing a singleset of scores a one way chi squareor G test can show the level of probability that the score frequencies differ non ran domly from Class to Class Theconfidence limits of each frequency see footnote8 Section 5 can show which particular frequencies differ signifi cantly from each other e g thenumber of Lichen Class 3s versusthe number of Class 2s If oneor morefrequency isnear zero asisoften the case with Score4s or 5s the confidencelimits will indicate whether it differs from zero by morethan random chance For comparing two sets of scores a two way chi square or G test can beused Typically this would beto comparetwo sites Either test can also be used for comparing a single set of scores with a pre defined target distribution An example might beto compare observed scores with a desired distribution of scores such asfrom

Download Pdf Manuals

Related Search

Related Contents

Manual del usuario de la L355  Sun Noise Reduction Kit Installation Instructions for Sun Ultra 30  CATALOGO DA51.2. 2007-2008 MICROMASTER 420  

Copyright © All rights reserved. Failed to retrieve file