United States Department 01 Snag Densities in Old-Growth Stands Agriculture on the Gasquet Ranger District, Six Service

Pacific Southwest Rivers National Forest, California Forest and Range Experiment Station

Research Paper Thomas M. Jimerson PSW-196 Jimcrson, Thomas M. 1989. Snag denslllcs in ald-growth slands an the Gasquet Ranger District Six Rivers Nalional Forest. California. Res. Pa~erPSW-196. Berkelev... CA: Pacific Swthwest Forest and Range Experiment Slation, Forest Sewice, U.S. Depan- men1 of Agriculture: 12 p.

Baseline levels for densities of snags (standing dead trecs) were determined in undis- turbed old-growth stands on the Gasquet Ranger District. Six Riven National Forest. California. Snag species, number, diameter at breast height (d.b.h.), height, cavity type. cavity use, decay class, and snagorigin wererecordedon 317plots overe.2-yearperiod. The 2121 snags ~eeo~dedconsisted of PseudDlsugo menziesii (36 percent). Abies concolor (29 perecnt), Abics mgnifica var. sharlemis (7 percent), Pinw lomberlinm (4 percent), Ch- mecyporis laws on in^ (6 percent), hardwmdr (10 percent), and other conifers combined (7 percent). Snags werecategorized as large, medium or small. Large snags were charauer- ized as snags 2 20 inchcs d.b.h. and 250 feet tall, the medium snsgs as snags 220 inches d.b.h. and 20-50 feet tall and lhe small snags as all other snags > 5 inches d.b.h. and 2 1 fmt raU that did not qualify as largeormedium snags. Thc snag densities for each category were then determined for each of the following conifer series: Lilhocorpus dem$7oralPseu- dolrugnmenricsii.Chomnccyparis Inwsoninno.Abiesconcolor and the Abiesmgnifc~var. shanenris. The snag densities for all four serics combined in cach category were 2.07 large snags!acre, 1.94 medium mags!acre and 20.67 small snags!acre.

Relrievol Term: mags, old-growth, , California, Pacific Nonhwest

The Author:

THOMAS M. JIMERSON is a zone ecologist with lhc Six Rivcrs National Forest in Eureka. California.

Acknowledgments:

'lhlr prpr reporrr a study sponsoreJ by lhc Su Rivers N=ttunal Forcsl un he G~iquct Ranger D~rlrict.in nanhue51 Cslifomla. I thak lhc followtng lndividulls fur lhclr arrlr- wnce in the rady: Thomas Q~mnand I'hdip Ron-fordala mllwuon: David Solts. David Diu, C. Jahn Ralph and Rruce B. Bingham-for study dcrign and data forms: .Michscl L. Morrison, Paula K. Kleintjes. James A. Baldwin, and Barbara H. Allen-for manuscript review.

Publisher:

Pacific Southwest Forest and Range Experiment Station P.O. Box 245, Berkeley, California 94701

July 1989 Snag Densities in Old-Growth Stands on the Gasquet Ranger District. Six Rivers National Forest. California

Thomas M. Jimerson

CONTENTS

.. In Brief ...... 11 Introduction ...... 1 Study Area ...... 1 Methods ...... 2 Results ...... 3 Study Area (All Series Combined) ...... 3 Conifer Series ...... 5 Discussion and Conclusions ...... 10 References ...... 11 IN BRIEF.. .

Jimerson, Thomas M. 1989. Snag densities in old-growth gory were then determined for each of the following conifer stands on the Gasquet Ranger District, Six Rivers Na- series: Lithocarpus densifZoralPseudorsuga menziesii, Cha- tional Forest, California. Res. Paper PSW-196. Berkeley, maecyparis lawsoniana, Abies concolor and Abies magnifica CA: Pacific SouthwestForest and RangeExperiment Station, var. shastensis. In general, snag densities follow a moisture or Forest Service, U.S. Department of Agriculture; 12 p. elevation gradient or both. Snag densities were found to vary by category and conifer series; however, this difference was not Retrieval Term: snags, old-growth, forest management, Cali- statistically significant for all categories. The snag densities for fornia, Pacific Northwest all four series combined in each category were 2.07 large snags/ acre, 1.94 medium snagslacre and 20.67 small snagslacre. Snags (standing dead ) andlogs arean importantcompo- The total number of snags recorded was 2121. Pseudorsuga nent of forest wildlife habitat and are considered a distinctive menziesii comprised 36percent of the total, with Abies concolor characteristic of old-growth . Many of the terrestrial contributing29percent,Abiesmagnificavar.shastensis7percent, vertebrates found in forests live or depend on snags and logs. A Pinus lambertiana 4 percent, Chamaecyparis lawsoniana 6 knowledge of baseline snag densities is required in order to percent, hardwoods 10 percent, and other conifers combined 7 manage this important resource. percent. Snag characteristics are presented for each conifer This study determined baseline levels for snag densities in series and species. Pseudotsuga menziesii had the highest rep- undisturbed old-growth stands on the Gasqoet Ranger District, resentation of snags and the best hard-to-soft snag ratio 40:60 Six Rivers National Forest, California. Stratified random sam- which indicates a constant rate of decay and suggests that pling with a nested sub-plot design was used to sample snag Pseudorsuga menziesii snags should be retained over other species, number, diameter at breast height (d.b.h.), height, species. cavity type, cavity use,decay class,andsnag origin on 317 plots These data will be combined with data on use by cavity- throughout the study area over a 2-year period. nesting birds to develop snag retention guidelines for the Six Snags were categorized as large, medium or small, based on Rivers National Forest. their importance to wildlife. The snag densities for each cate- - - -- INTRODUCTION

Id-growth conifer forests in the Pacific Northwest are many of the National Forests in Califomia, information on valued economically for their timberand as wildlife habi- baseline snag density and of snag management effects on snag- tat (Harris 1984). Snags (standing dead trees) and logs are an dependent wildlife is inadequate. To date, few studies examin- important component of this wildlife habitat (US. Dep. Agric. ing snag densities by conifer series and vegetation types have 1986). and large snags are considered as one of the distinctive been attempted. features ofold-growthforests (Franklinandothers 1981). Ninety This paper reports basetine snag densities for the primary percent of the terrestrial vertebrates that exist in our forests live conifer series of the Gasanet Raneer- Dismct. Six Rivers Na- or depend on such (Franklin [in press]). tionalForest.anddescribes snag characteristicsby conifer series Snags provide habitat for foraging, nesting, resting or cover for and species. The data generatedby this study will be usedby Six many species of wildlife (Thomas and others 1979). Character- Rivers National Forest personnel to develop snag retention istics of snags, including state of decay, density, size, and guidelines for the Forest. species, influence their use by wildlife (Mannan and others 1980,Maser andothers 1979,Raphael 1980). For example, soft snags are most often used for nesting whereas hard snags are most often used for foraging. In addition, the diameter atbreast STUDY AREA height and height of snags determine which species will use a snag for nesting (Thomas and others 1979). Many research programs have focused upon old-growth The study was conducted on the Gasquet Ranger District in Douglas-fix forests because they are a major vegetation type in Del NorteCounty about 8 miles (13 kilometers) eastof Crescent the Pacific Northwest (Franklin and others 1981, Mannan and City Vg. I).Itincludes about 249,000 acres (101,000 hectares) others 1980, Cline and others 1980) and particularly because of primarily coniferous forest (U.S. Dep. Agric. 1976). The intensive timber management tends to eliminate older stands, Gasquet Ranger District is characterized by warm dry summers eliminate dead and dying trees, and remove snags as fire and and cool wet winters. Itranges inelevation from 100 to6000 feet safety hazards (Thomas and others 1979). These management (30-1800 meters). strategies conflict with the goals of managing snags as wildlife habitat. Clear-cut , , and short harvest rotation periods adversely alter or eliminate the potential of a timber stand to retain or produce snags (Mannan and others 1980). The National Forests in the Pacific Southwest Region have established guidelines for the retention of snags (table I).These guidelines are based on the work of Thomas and others (1979) intheBlueMountains of 0regon.Theguidelinesarenot specific to California and should be supplemented with site-specific information when possible. For example, the Tahoe National Forest modified Thomas and others' (1979) guidelines on the basis of the work of Raphael and White (1984). However, for

Table I-USDA Forrsl Service guidelines for the refenlion ofsmgs in lhe Pacific Soulhwcst Region

Direction for snae-deoendent rnecies

Withii each timber canpament, provide, maintain, and manage. to the extent possible, for an averago of 1.5 snags per acre with the following specifications:

(2) 0.3 mags per acre that are greater lhan 24 inches d.b.h. and higher than 20 feet.

Other requirements for snag disrribution (e.g.. hard-to-soft ratios) will be included in Foorertwide standards and guidelines, prescriptions, or manage- ment area direction in the Forcrt plan. Flgure I-The Six Rivers National Forest in Northern Californiawith the study area aosshatched.

USDA Forest Service Res. Paper PSW-196. 1989. The vegetation in the study area includes four conifer series (fig. 2, table 2). 1) The Port-Orford-Cedar (Chamaecyparis lawsoniana (A. ~urr.)~ar~.) series is located along the stream METHODS bottoms. 2) The tanoak/Douglas-fu (Lirhocarpusdensflora (H. and A,) Rehd.lPseudotsuga menziesii (Mirb.) Franco.) series begins at the bottom of the slopes and continues upslope to This study was conducted in conjunction with the USDA approximately 4000 feet (1200 meters). 3) The white fu (Abies ForestSewice'secosystem classification program (Allen 1987) concolor [Gord. and Glendl.] Lindl.) series replaces the tanoak/ being conducted on the Six Rivers National Forest. Late seral- Douglas-fir series above 40M) feet (1200 meters), and 4) the red stage stands (old-growth) were selected as study sites. Old- fir (Abies magnifica A. Mun. var. shastensis Lemmon) series growth stands were located through a review of timber type replaces the white fir series at the top of the highest mountains maps (U.S. Dep. Agric. 1976) and verified through analysis of (fig. 2). Small pockets of Jeffrey pine (Pinus jeffreyi Grev. and aerial photos and ground reconnaissance. Sample stands were Balf.), lodgepole pine (Pinus conlorta Dougl.), and knobcone selected by using the methodology of Pfister and Arno (1980). pine (Pinusattenuata Lemmon) are found throughout the study This methodology was modified to include only stands older area, but were not sampled because of their limited extent and than 150 years that displayed all or most of the characteristics lack of management.

Table 2-Approximore acreage for each conifer series in the study oreo Conifer series Approximate acres Pct

Tanoak/Douglas-fir 210,000 84

Port-Orford-Cedar 5.000 2

While fir 27.000 11

Red fir 2,000 1

Olher' 5,000 2

Total 249.000 100

'Contains Jeffrey pine, lodgepole pine, and knobmne pine.

Figure 2-Relative positions of each conifer series within the study area.

USDA Forest Service Rer. Pacer PSW-196. 1989. defined by Franklin and others (1981) and the Old-Growth >1foot tall. In addition, snags were divided into three descrip- Definition Task Group (1986). Within selected stands, stratified tive categories (large, medium, and small) on the basis of guide- random sampling was used to select the plot locations following lines on snag retention, from the Pacific Southwest Region, the methodology of Noon (1981). which included placing non- USDA Forest Service, and the literature. Large snags are 2 20 overlapping plots wilhin all distinct habitats within a stand. In inchesd.b.h.andz50feettall(Mannanandothers 1980,Bulland watershedswith fewremaining old-growth stands, allremaining Meslow 1977), medium snags are? 20 inches d.b.h. and 20-50 stands were sampled. In designated wilderness areas in which feet tall,andsmallsnagsare> 5inches d.b.h. anddonotmeet the the majority of the area is in old-growth stands, transects were requirements of large or medium snags. located perpendicular to the slope with plots located every 300 Presence of excavated cavities in snags was considered a sign feet (80 meters). Approximately 2percent of theacreagein old- of previous or current use of snags by wildlife. Use of natural growth stands in the study area was sampled. cavities in snags by wildlife was determined by direct observa- To determine the appropriate plot size, the pertinent literature tion of animals in cavities orby inference. Use by inference was on snag sampling was reviewed. The review indicated that plot determined by an accumulation of scat or nesting material, or sizes generally fell into three categories: large, medium, and scratchings, scraping, clearing, or exposing of soil or bark. small. Thesecategories generally reflected the objectives of the The statistical program SPSS/PC+ (Norusis 1986) generated particular study. The large plot category included the 21-acre statistics mean, frequency, mode, density, standard deviation, (8.5-hectare) plots of Raphael and White (1984), 7.5- to 37-acre and standard error. (3-15 hectare) plots of Gale (1973), 25- to 37-acre (10-15 hectare) plots of Cunningham and others (1980), and the 12-acre (5-hectare) plots of Marzluff and Lyon (1983). The medium plot category included the plots of Brush andothers (1983) K2.5- RESULTS acre (l-hectare) plots], Morrison and others (1986) r0.25-acre (0.1-hectare) plots], Carmichael and Guynn (1983) [0.25-acre (0.1-hectare) plots], and Cline and others (1980) [1.2-acre (0.5- hectare) plots]. The small plot category included several studies Study Area (All Series Combined) using a 0.1-acre (0.04-hecme) plot (Cline and others 1980, Mannan and others 1980, McComb and Muller 1983, Moriarty Data werecollectedat 317plots, which contained2121 snags. and McComb 1983, Morrison and others 1986, Swallow and Snag densities for the total study area were highest for small others 1986). snags, followed by large snags and medium snags (table 3). The literature review indicated that the 0.1-acre plot was the Snags were composed primarily of conifer species; Pseu- preferred size for this type of study. However, because the dotsuga menziesii (36 percent) was the dominant species, fol- studies that used the 0.1-acre plot were not conducted in north- lowed by Abies concolor (29 percent), Abies magnifica var. west California, pre-study sampling was conducted. This sam- shastensis (7 percent), Chamaecyparis lawsoniana (6 percent), pling involved counting snags within plots of various sizes. We and Pinus lambertiana (4 percent). Hardwoods made up 10 determined that adequate numbers of snags would be included percentof thesnags, most of which were foundin the small-snag in each plot if two nested study plots of 0.5 and 0.25 acres were category (table 4). used. The plot size of 0.5 acre was used to examine snags 220 The small-snag category had the largest number of trees inchesindiameteratbreast height(d.b.h.)and?5Ofeettall(large (1638). The mean d.b.h. was 14 inches with amean height of 18 snags). The 0.5-acre plot size was selected instead of the stan- feet. The small-snag category was dominated by Pseudolsuga dard 0.1-acre plotsize because of thelow densities oflarge snags in the study area and their importance to wildlife (Bull and Meslow 1977, McClelland and Frissell 1975). A plot size of Table 3-4~18~mry for all conifelseries combined, on 317 plols 0.25 acre was used to examine all snags with a d.b.h. >5 inches and >1 foot tall (small snags) that didnot meet the requirements of large snags. The following characteristics were recorded for all snags: (1) species; (2) d.b.h., measured with a diameter tape to the nearest inch (forsnags<4.5ft. tall, d.b.h. wasestimated); (3) decay class (five decay classes according to Cline and others 1980); (4) height, measured directly with a tape or estimated with a Relaskop to thenearest foot; (5) cavities (presence of excavated nestcavities and presenceand useofnaturalcavities inroot, butt, or midbole of ); and (6) origin (snag originated in previous or current stand of timber, or tree was cut). Decay classes 1and 'Lurge snags are 2 20 inches d.b.h. and 250 fcet tall. 2 wereconsidered hard snags, whereas decay classes 3 through Medium snags are 220 inches d.b.h. and 20-50fcet tall. 5 were considered soft snags. For the purpose of this study, 'Small snags are all other snags that do not meet the requirements ior snags were defined as standing dead trees >5 inches d.b.h. and large or medium snags.

USDA Forest Service Res. Paper PSW-196. 1989 Table specis composilion summary for ihe srudy area, N=2121

Species large snags Medium snags Small snags Total No. Pcr No. Pcr NO. ~ct NO. ~cr +PseudoLruga menrierii 155 47 79 51 521 32 755 36 Chamecypnrir low son in^ 29 9 8 5 88 5 125 6

Abies concolor 55 17 39 25 502 31 596 29

Abies mgnifien var. shnrrenrir 27 8 7 4 110 7 144 7

Pinw lamberrinno 44 13 8 5 41 3 93 4

Pinw monricolo 11 3 6 4 38 2 55 3

Libocedrw decurrenr 3 1 0 0 17 1 20 1

Unknown conifer 0 0 0 0 65 4 65 3

Hardwood 0 0 5 3 209 13 214 10

Others 4 1 3 1 47 3 54 2

menziesii. Abies concolor, and hardwoods. The large snag decreasing rapidly in decay classes 3 through 5. This trend category had the second largest number of trees; it contained328 continued for the major conifer species with the exception of snags with a mean d.b.h. of 32 inches and a mean height of 94 Chamuecyparis lawsoniana fig. 3). In the small-snag category feet. It was dominated entirely by conifers, with Pseudotsuga the proportion of snags in decay class 2 was again the highest. menziesii. Abies concolor, and Pinuslamberdana as the primary Hardwoods had the same proportion in each decay class, ap- species. The medium-snag category had the fewest snags (155). proximately 20 percent fig. 4). The ratio of hard to soft snags Themean d.b.h. was 26inches and themean height was 37 feet. indicates a greater proportion of snags are found in the hard The medium-snag category was primarily dominated by the category. The ratio of hard to soft snags varies by species. conifers Pseudotsuga menziesii and Abies concolor. Pseudotsuga menziesii has a40:60ratio of hard to soft,followed The modal decay class for all snag categories in the study area by Chamuecyparis lawsoniana 65:35, Abies concolor 65:35, was 2 (hard snags). The largest proportion of large and medium Abiesmugnijica var.shastensis65:35,Pinuslambertiona 80:20, snags was found in decay class2, followed by decay class 1 and and hardwoods 45:55.

CONIFER SPECIES '-C PSME " CHLA + ABCO * ABMAS -* PlLA -'-PIM03 4- HARDWOOD 1

1 2 3 4 6 s 1 2 3 4 6 DECAY CLASS DECAY CLASS

Figure 3-Percent of large and medium snags by decay class and Figure &-Percent of small snags by decay class and species (PSME species (PSME = Pseudotsuga menziesi, CHLA = Chamaecyparis = Pseudotsuga menziesii, CHLA = Chamaecyparis lawsoniana, ABCO lawsoniana, ABCO = Abies concolor, ABMAS = Abies magnifica var. = Abies concolor, ABMAS = Abies magnib var. shastensis. PllA = shastensis. PILA = Pinus lamberfiana). Pinus lamberliana, PIM03 = Pinus monticola).

4 USDA Forest Service Res. Paper PSW-196. 1989 Table 5--Chnmderirlics ofsnags in :he lononklDouglaslfir (Lithocarpus Table 7-Charoclcrir~icsof~~g~infhePor:-Orford-Ce&~(Chmaeqp~~s dcnsiRaia/Pseudotsuga menziesii) series in 166 plols lawaoniana) series in 56 plors

Mean Standard Standard Mean Standard Standard Snag size Snags D.b.h. / Height Density deviation error Snag size Snags D.b.h. I Height Density deviation error in. 3. snagdocre in. /I. smgs/acre / Large' 1 130 34 100 1.57 1.11 0.09 / Medium' 41 27 37 2.86 0.99 0.13 / Medivm + Large 127 31 74 5.93 6.59 0.88 ( 13.64 3.25 0.25 I m 350 5 20 25.00 5.09 0.68 'Large snags are 220 inches d.b.h. and 250 feet tall. 'Large snags are220 inches d.bb. and 250feertdl. lMedium snags arc 220 inches d.b.h. and 20-50 feet tall. lMedium snags are 220 inches d.b.h. and 20-50 feet tall. 'Small snagr are dl other snags that do not meet the rcqvirements for 'Small snags arc all other snags that do not meet the requirements lor large or medium snags. large or medium snags.

Conifer Series The Chamaecyparis lawsonianaseries contained477 snags in 56 plots. The small-snag category had the highest density, 25.00 The Lilhocarpus densifloralPseudotsuga menziesii series snags/acre, followed by the large- and medium-snag categories, contained 744 snags in 166 plots, most of which were found in 3.07 and 2.86 snagslacre (table 7). Pseudotsuga menziesii, thesmall-snagcalegory.Densities varied from 13.64 snagslacre Chamaecyparis lawsoniana and Abies concolor were the domi- for small snags, to 1.57 and 1.16 snagslacre for the large- and nant snag species (table 8). The modal decay class for the large- medium-snag categories (table 5).Pseudotsuga menziesii and and small-snag categories was 2, whereas the medium-snag hardwoods were thedominant snag species (table6). Themodal category hadamodeof3 or4 (table 7). TheAbiesconcolorseries decay class for large snags was 2, whereas the medium- and contained 596 snags in 70 plots. The small-snag category had the small-snag categories both had modes of 3 (table 5). highest density, 26.86 snagslacre, followed by medium- and

Table 6--Composirion of snags in the ronoo~Douglns-fir(Lithocarpus densiflorflseudotsuga menziesii) series N=744

Species / Large snags / Medium snags / Small snags / Total No. Pd No. Pcr No. Pcl No. Pcr

Pseudorsuge menriesii

Chomnecyparir lowsoninnn

Abier concolor

Abies mngnificn var. sharrenrir

Pinus lomberliam

Pinu. monlicoln

Libocedrus decwrenr

Unknown conifer

Hardwood

Others

USDA Forest Service Res. Paper PSW-196. 1989. Table Wornposition of smgs in the Port-Orford-Cedar (Chamaecyparis lawsoniana) series N=477

Specics Largc snags Medium snags Small snags Total

Psewlotsugn menziesii

Chnmnecyparir lnwsonio~

Abies concolor

Abies mngnifica var. shnslensir

Pinus lambertiona

Pinw monlicolo

Libocedrus decurrenr

Unknown ConiCer

Hardwood

Others

large-snag categories, 2.57 and 2.31 snagslacre r able 9). Abies snagslacre, followed by medium- and large-snag categories, concolor and Pseudotsuga menziesii were the dominant snag 3.36 and 2.48 snagslacre (table 11). Abies concolor and Abies species (table 10). The modal decay class for all three categories magnifica var. shastensis were the dominant snag species (table was 2 (table 9). 12). The modal decay class for all three categories was 2 (table The Abies magnifica var. shastensis series contained 304 11). snagsin25plots. Thesmall-snagcategory hadadensity of40.32 The Chamaecyparis lawsoniana series had the highest snag

Table 9--Chnroclerirrics of snags in the whilefi (Abies concolor) seri~sin Table 112hnroctcrinics ofs~gsin the Shnsln redfir (Abiesmagnifica "or. 70 plots shastensis) series in 25 plots

Mean Standard Standard Mean Standard Standard Snag size Snags D.b.h. Hcight Density deviation enor Snags D.b.h. Height Density deviation error I I in. p. snapslacre I in. fl. snagslocre Large' 81 31 88 2.31 1.77 0.21 Large'

Medium2 45 25 36 2.57 0.93 0.11

Medium + Medium + Largc 126 29 69 4.89 5.13 0.61 1 !.aW.e Small' 470 14 19 26.86 5.07 0.61

'Lrgc rnlgr ire 120inches J.b.h. anJ 2 50 feel nll. 'Large snags are 2 20 inches d.b.h. and 2 50 feet tall. '\ledlum wsgr are >2U inchcs dbh. and 20-50 ice, vsll. 2Medium snags are 2 20 inchos d.b.h. and 20-50 feet tall. 'Small mAcr- arc aU other stucs- h31do "01 meel the rau:rerncntr fcr 'Small snsgs are all other snags that do not meet the requirements for large or medium snags. large or medium snags.

USDA Forest Service Reo. Paper PSW-196. 1989. Table lO-Compnririon of smgs in the whilefir (Abier mnmlor) series N=596

Pscudouuga memicsii 20 25 17 38 91 19 128 21

Cha"~ecypnr!~lawsonkm 11 0 0 5 1 6 1

Abics concolor 41 51 24 53 317 67 382 64

Abies magnifica var. shasrenrir 4 5 0 0 4 1 8 1

Pinw lambeniam 7 9 12 4 1 12 2

Pinu monlicola 6 7 2 4 11 2 19 3

Liboce&w dccwrenr

Pinw nllanuala

Unhown Conifer Hardwood Others

Table U-Comparbion ofsmgs in the Shnsla redfir (Abies magnifica vnr. shastensir) series N=304

Species / Large snags Medium snags Small snags Toral No. Pcf No. Pa No. Pcr No. Pcr

Pscudotruga memiesii

Chamarcyporir lowsonia~

Abies concolor

Abies magnifico var. shasfcnrir

Pinvr lombeniana

Pinw monlicola

Liboce&vr decurrrnr

Unknown Conifer

Hardwood

Picen brewcriam

USDA Forest Service Res. Paper PSW-196. 1989 Table 13---Comparison ofsnag denxilies by conifpr series

Series (~ar~c/ Medium / Large t Medium ( Small snogsi',cre

Lilhocarpw densifmi

Pseudolsugn menriesii 1.57 1.16 2.72 13.64

Chamnecypr* lawsoninno 3.07 2.86 5.93 25.00

Abies concolor 2.31 2.57 4.89 26.86

Abies magnificn var. slvrrlensir 2.48 3.36 5.84 40.32

density for the large-snag category, the Abies magnifica var. Snags varied in size and decay class according to species and shastensis series had the highest snag density in the medium- d.b.h. Conifers dominated the large- and medium-snag catego- snagcategory, the Abies magnifica var. shastensis series had the ries. Hardwoods began to appear in the medium-snag&eg&. highest density in the small-snag category (table 13). and codominate in the small-snag category. Decay classes The proportion of large to medium snags in each conifer series generally followed beecategories. Mostof the largesnags were and decay class indicates that the majority of snags within the found in decay class 2, whereas the number of snags in decay study area are included in decay class 2, followed by decay class 3 increased in the medium category, and decay class 4 classes 1,3,4 and 5 (fig. 5). The small-snag category is more appeared more frequently in the small-snag category (table 14). evenly distributed, with decay class 2 having a slightly higher Pinus lambertiana conuibuted the tallest snags, with a mean proportion of snags O%g. 6).The proportion of hard to soft snags height of 107 feet in thelarge-snag category. It was followed by varied by conifer series in the large- and medium-snag catego- Pseudotsuga menziesii (94 feet), Abies magnifica var. shasten- ries. The Chamaecyparis lawsoniana series had a ratio of hard sir (90 feet), Chamaecyparis lawsoniana (89 feet), and Abies to soft snags of 60:40. TheLithocarpus densifloralPseudotsuga concolor (87 feet) (table 14). Snag height increased by diameter menziesii series had a ratio of 7525, while the Abies concolor class except for the 40-inch d.b.h. class (table 15). Decay class series had aratio of 80:20 and the Abies magnifica var. shasten- 2 was the modal decay class in all categories except the 40-inch sis series had a ratio of 85:15. d.b.h.category,forwhich themodaldecay class was5 (tableI5).

00 1 2 3 4 6 DECAY CLASS DECAY CLASS

Figure 5-Percent of large and medium snags in each decay class by Flgure GPercenl of small snags in each decay class by coniler serles conifer series (LIDEBPSME = Lirhocarpus densiflora-Pseudorsuga (LIDE2-PSME = L8rhocarous densillora-Pseudorsuoamenzies~i. CHLA menziesii, CHLA = Chamaecyparis lawsoniana, ABGO = Abies con- = ~hamaecyparis1awsor;iana. ABCO = ~biesconcilor, ABMAS = Abies color, ABMAS = Abies magnifica var. shastensis). magnifica var. shastensis).

USDA Forat Service Res. Paper PSW-196. 1989. Table 14Snag species descriptions, by snog category

Large snags Medium snags Small snags Mean Decay Mean Dccay Mean Decay Species Snags D.b.h. / Height class Snags D.b.h. / Height class Snags D.b.h. I Height class

Abies concolor

Abies mngnflca var. shasfensis

Pinw lnrnbefliana

I-lardwoods

Arburlrs menziesii

Coslanopsis chrysolepis

Lilhocnrplrs densiflora

Quercw clvysolepis

Table 1Mmgdescripfions, by dinmeler class (d.6.h.)

Diameter (D.b.h.) I Mean 1 Decay class Class 1 Range N Height 1 2 3 4 5

I I "This diameter class normally ranger from 1 to 5.9 inches. In this study no snags were measured below 5 inches diameter.

USDA Forest Setvice Res. Paper PSW-196. 1989. regenerates under the forest canopy and is distributed across the forest floor. Themost tolerant series is the Abiesmagnifica var. DISCUSSION AND CONCLUSIONS shasfensis. Reproduction takes place in thick clumps in small or large openings or is spread thinly throughout closed canopy stands. Thesevarying reproductive strategies lead to the differ- ences noted in small-snag densities (fable 13). Snag densities for old-growth stands on the Gasquet Ranger The composition of snag species was determined for the 2121 District vary by snag category and by conifer series. In general, snags recorded. Pseudofsuga menziesii was the most important densities were highest in the small-snag category, followed by snag contributor because of its dominance or codominance the large- and medium-snag categories (fable 13). The small- within allconifer series except the Abiesmagnificavar. shasten- snag category is expected to have the highest density because sis (Sawyer and Thomburgh 1977, Sawyer and others 1977). there were more small trees than large trees in the study stands. Pseudofsuga menziesii was followed by Abies concolor, which Also, because of competition for growing space, mormty is had high numbers and was also represented in all four series. higher in thesmall-snagcategory. The lowerdensity found in the Abies concolor was found in the higher-elevation portions of the medium-snag category seems to be related to the length of time Lifhocarpusdensifora/Pseudorsuga menziesiiseries,thelower- since death and the decay rate of each species. Snags in the elevation portions of the Abies magnifica var. shasfensis series, medium category are220 inches d.b.h. and20-50 feet tall. Snags in which it is a codominant conifer, in limited numbers in the included in this category could result from 1) the death of Chamaecyparis lawsoniana series, and throughout the Abies medium sized trees, 2) large snags with broken tops, or 3) the concolor series in which it is the dominant conifer (Sawyer and decay of large snags and associated reduction in size. The mean Thornburgh 1977). Hardwoods had the next largest number of d.b.h. for the medium-snag category is lower than the mean snags; they are found primarily in the Lithocarpus densif[oral d.b.h. for the large-snag category (fable 3),thereby indicating Pseudofsuga menziesii series, in which they often codominate that all three sources are probable contributors, with source 1 with Pseudotsuga menziesii (Sawyer and others 1977). Hard- (death of medium sized trees) as the primary contributor. species were important snag contributors in the small- In general, snag densities of conifer series lollow a moisture snag category only. Abiesmagnifica var. shastensiscontributed or elevation gradient or both. The Chamaecyparis lawsoniana the next largest number of snags. It was found in the higher series has the highest overall snag densities. This series is elevation portions of the Abies concolor series, in which it can located near streams with perennial water, which enables it to be a codominant, and throughout the Abies magnifica var. mainlain a high density of stems.' The Abies magnifica var. shasfensisseries in which itis thedominantconifer (Sawyerand shastensis series had the second-highest snag densities. It is Thornburgh 1977). Abies magnifico var. shastensis was fol- located at the tops of the mountain peaks, on the upper third of lowedclosely in snag contribution by Chamaecyparis lawsoni- the slope; here moisture and stand density (Davis and Johnson ana. Chamaecyparislawsoniana was found primarily in its own 1987) are higher than the locations of all other series except the series, which is of limited geographic extent in the study area Chamaecyparis lawsoniana series. The Abies concolor series (table 2). Thesmallest numbers of snags contributed by a major has the thud-highest snag density. It is located just below the conifer species came from Pinus lamberfiana (fig. 3). It was Abies magnifica var. shasfensisseries, in the upper two thuds of found in theLithocarprrsdemiflora/Pseudorsugamenziesiiseries the slope, where the amount of moisture drops with decreasing in which it can be a codominant conifer (Sawyer and others elevation along with stand density. The Lithocarpus densiforal 1977) and in the Chamaecyparis lawsoniana series in which it Pseudofsuga menziesiiseries is located in thelower one third of is a minor forest component. thes1ope.Ithadihelowestsnagdensityandalso theloweststand The processof change from decay class 1 through decay class density (Davis and Johnson 1987). 5 is related to the decay rate of wood. The principal organisms Snag densities also vary by category within a conifer series. responsible for decay are fungi, , and (U.S. Dep. These differences are related to the reproductive strategies Agric. 1974). The rate of decay is influenced by climate, employed by the dominant conifer species. The Lithocarpus elevation, aspect, tree size, percent sapwood and tree species densifloralPseudouugamenziesii,Chamaecyparislawsoniana, (Kimmey 1955). A resistance to decay rating of the principal Abies concolor, and Abies magnifica var. shastensis conifer conifer species shows that Chamaecyparis lawsoniana is rated series follow a reproductive tolerance gradient (Sawyer and as resistant or very resistant to decay. It is followed by Pseu- Thomburgh 1977). Reproduction in the Lifhocarpus den- dofsuga menziesii and Pinus lamberfiana, which are moderately sifloralPseudofsuga menziesii series is the least tolerant to resistant to decay, whereas Abies concolor and Abiesmagnifica shading.Within thisseries,regeneration takesplaceprimarily in var. shasfensisare slightly or nonresistant to decay (U.S. Dep. the gaps left by falling trees or in openings created by fire or Agric. 1974, Kimmey 1955). Decay state is very important to logging. Next in tolerance is the Chamaecyparis lawsoniana cavity-nestingbirds.Excavationofcavities does notbeginuntil series,in which regeneration takesplaceunder the forestcanopy the snags reach decay class 2 (Mannan and others 1980, Cline or in openings. TheAbies concolor series is next in tolerance. It and others 1980)probably because thesnags in decay class 1 are too hard for excavation (Mannan and others 1980). Acomparison of percent snags in each decay class by conifer 'Unpublished data on fie at Six Rivers National Forest, Eureka, Cali. series indicates small differences between series (figs. 5 and6)

10 USDA Forest Service Rer. Paper PSW-196. 1989 which are due to the different decay rates and species compo- sition of each series, whereas an examination by conifer species indicates much larger differences Wg. 3). Chamaecyparis REFERENCES lawsoniana is rated as resistant to decay (U.S. Dep. Agric. 1974). It has its lowest proportion of snags in decay class 1 and its highest proportion in decay class 2 (fig. 3). After reaching decay class 2, thedecay rate seems to increase (fig.3). The ratio Allen, Barbara H. 1987. Ecological type elassificatio,n far California: the of hard to soft snags is 21, indicating a slow decay rate during Forest Service approach. Gen. Tech. Rep. PSW-98. Berkeley. CA: Pacific decay classes 1 and 2, and a more rapid decay rate in classes 3 Southwest Forest and Range Experiment Station, Forest Service. U.S. Dc- through 5. Pseudotsuga menziesii is rated as moderately resis- parunent of Agriculmre; 8 p. Brush, Timothy; Anderson. Bcnin W.; Ohman. Roben D. 1983. Habitat tant to decay (U.S. Dep. Agric. 1974). It had the most even selection related toresourceavaiiability amongcavity-nesting birds.In: distribution throughout all decay classes (fig. 3), and the best Davis,Jerry W.; Goodwin, Gregory A.; Ockeniels,Richard A,. tech. caords. ratio of hard to soft snags, 40:60. This ratio indicates a constant Snag habitat management: proceedings of hesymposium; 1983 June7-9; rate of decay across all decay classes, suggesting that Pseu- Flagstaff, AZ. Gen. Tcch. Rep. RM-99. Fon Collins. CO: Rocky Mountain dotsuga menziesii should be selected over other snag species. Forest and Range Experiment Station, Forest Service, U.S. Depament of Agricubure; 88-98. Pinus lambertiana is also rated as moderately resistant to decay Bull. Evelyn L.; Meslow. E. Chades. 1977. Habitat requirements of the (U.S. Dep. Agric. 1974). Proportionately, the largest number of pileated woodpccker in northeastern Oregon. Journalof Forertry75:335- its snags appear in decay class 1. The proportion shows a 337. moderate decrease in decay class 2, and a large drop in decay Carmiehael, D. Bredt, Jr.; Guynn, David C., Jr. 1983. Snag density and classes 3 through 5. The ratio of hard to soft snags is4: 1, which utilization by wildlife in the upper piedmont of South Caroiina. In: Davis.Jerry W.; Goadwin. Gregory A,; Ockenfels, Richard A,. tech. coords. indicates that once the Pinus lambertiana snags reach decay Snag habitat management: proceedings of the symposium; 1983 June 7-9; class 2, they decay rapidly until they disappear from the stand Flagstaff, AZ. Gen. Tech. Rep. RM-99. Fort Collins, CO: Rocky Mountain (fig3). The results of this study indicatethatPinuslambertiana Forest and Range Experiment Station. Forest Service. U.S. Department of islessresistanttodecay thanPseudotsugamenziesii. BothAbies Agriculture; 107-110. conco~orand~biesmagnificavar,shastens~sarerat~ass~ighl~yCline, S.P.; Berg, A.B.; Wight. H.M. 1980. Snag characteristics and dynam- lcs in Douglas-fir forests, western Oregon. Journal of Wildlife Manage- or nonresistant to decay (U.S. Dep. Agric. 1974). Both have .ment. . - ... 44:771-786~...... -.. smaller proportions in class and lheir highest Cunningham, James B.; Balda. Russell P.; Gaud, Wiiam S. 1980. Selection proportions in decay class 2 (fig.3). The ratios of hard to soft and useofsnags by secondary cavity-nesting birds of the ponderosa pine snags for both species are approximately 2: 1. The small propor- forest. Res. Paper RM-222. Port Collins. CO: Rocky Mountain Forest and tion of soft snags indicates a limited time spent in decay classes Rangc Experiment Station, Forest Sewice, U.S. Department of Agriculmre; .12. p. 3 through 5. Kimmey (1955) found that Abies concolor snags Davis, L. S.; Johnson, K. N. 1987. Forest management. 3d ed. New York: decay rapidly andarenot salvageableafter4 years. His iindings, McGraw-Hii Book Company. 790 p. along with the results of this study, indicate that these species Franklin. J.F. Bioioricai- Iceacies.- In: Land classifications based on vegetation:- have limited longevity as snags. appiiulions for resource mrnsgcment. I'ruccedingr of s symporum; 1987 Although snag age was not determined, the proportions of A'overnbcr 16-19; .Moirou, ID. Ocden, KT: Jnlemoun~inForen anJ R,ntc Experiment Station, Forest Service, U.S. Department of Agriculture. F snags in each decay class indicate that snag longevity isgreatest press]. in decay class 2 since the proportion of all snags with the Franklin. Jerry F.; Cromack. Kermit, Jr.; Denisan. William; and others. 1981. exceptionofPinuslambertianaishighest here. Decay classes3- Eeolagieal characteristics of old-growth Douglas-fir forests. Gen. Tech. 5 contain smaller, decreasing numbers of snags for all species. Rep. PNW-118. Ponland, OR: PaeificNonhwest Forest and Range Experi- In addition, snag longevity is influenced by size (Kimmey ment Station, Forest Service, U.S. Depanment of Agriculture: 48 p. 1955). For the large- and medium-snag categories, the propor- Ga1e.R. 1973.Snags,chainsawsand wildlife: oneaspectofhabitatmanage ment. Cal-Neva Wildlife: 97-111. tions of hard snags are much greater than soft snags for all Hanis, L.D. 1984. The fragmented foresl: island biogeography thcory and conifers. The data indicate that snags decay slower in the large the preservation of biotic diversity. Chicago, k University of Chicago and medium categories and there are fewer soft snags. The Press. 211 p. importance of this soft-snag component to cavity-nesting birds Kimmey. 1. W. 1955. Rate of deterioration of f1re.kilied timber in Califor- (Thomas and others 1979, Mannan and others 1980, Cline and nia. Washington, DC: U.S. Depment of Agriculmre; 22 p. Mannan. R.W.; Meslow, E.C.; Wight. H.M. 1980. Use of snags by birds in others 1980) indicates a need to maintain soft snags in higher Douglas-fir forests, western Oregon. Joumal of Wildlife Managemen1 ~roaortionsthan hard snags. 44(4):. . 787-797. - The implication of this study for managers is that snag Marzlulf, John M.;Lyon. L. Jack. 1983. Snags as indicators of habitat management must be site-specific because snag densities vary suitability for open nesting birds. In: Davis, Jeny W.; Goodwin. Gregory by conifer series and snag category, and snagsdecay at different A,; 0ckenfels.Richard A,, tech. cmrds. Snag habitatmanagement: proceed- ings of the symposium; 1983 June7-9; Flagstaff, AZ. Gen. Tech. Rep. RM- rates. 99. Fon Collimr. CO. Rocky Mountain Forest and Range Experiment Station. Forest Service. U.S. Depanment of Agriculture; 140-146. Maser. C.; Anderson, R.G.; Cromack. K.. Jr.. [and othenl. 1979. Dend and down woody material. In: Thomas, LW.. tech. ed. Wildlife habitat in managed forests-the Blue Mountains of Oregon and Washington, Agric. Handb. 553. Washington. DC: U.S. Depament of Agriculture; 78-95.

USDA Forest Service Rer. Paper PSW-196. 1989. McClolland, B. R.; FrirseU. S.S. 1975. Identifying forest snags useful for Pfister, R.D.; Arno, S.T. 1980. Classifying forest habitat types based on hoic-nesting birds. Journal of 73:414-417. potential climax vecclntian. Forest Science 26: 52-70. McCornb, w.c.; Mullcr, R.N. 1983. snag dcnsities in oid.growth and ~aihael.~.~.1980. ~iiizationofstanding dead trees by brecding birds nt second-growth Appalachian forests. Journal of Wildlife Management Sagehen Creek, Cnllfornia. Berkeley. CA: University of California, t95p. 47:376-382. Disscrtalion. Moriany,JbhnJ.;McCornb. William C. 1983.Thelongterm olTectof limber Raphael, M.G.; White. M. 1984. Use of snags by envity-nesting birds in the stand improvemcnt an snag and cavity densities In the central Appala. Sierra Nevada. Wildlife Monographs 86. chinns. In: Davis. Jerry W.: Goadwin, Gregory A:, Ockenfels. Richard A,. Sawyer. 1.0.: Thornburgh, D.A. 1977. Montane and subalpine vegetation of tech. caards. Snae habitat management: amceediinas of the svm~osium: the Klamath Mountains. In: Barbour, M.G.; Major. I., cds. Terrestrial 1983 June 7-9; ~lagstsff,AZ Gc;. Tech. kep. RM-99. Fort C&ins, CO: vegetation of California. New York: John Wiley and Sons, Inc.; 699.732, Rocky Mountain Farestand Range ExperimentStation, Forest Service, U.S. Sawyer. 1.0.: Thornburgh. D.A.; Griffin, J.R. 1977. Mixed evergreen forest. Deparunent of Agriculture; 40-44. In: Barbour. M.G.; Major, 1.. cds. Terrestrial vegetation of California. New Morrison. Michael: Dcdon. Mark F.; Yoder-Williams, Michael P. [and alhenl. Yok John Wiley and Sans. Inc.; 359-381. I986 Distribution 2nd obundlnreof mags in the Sagehen Creek Basin, Swallow. S.K.; Gutierrez, RJ.; Howard, R.A.. Jr. 1986. Primary cavily-site California Kc$.Sole PSW-389. Berkcicy, CA: Paciftc Sourhuest Porest seleclion by birds. Journal of Wildlife Managanent 50: 576-583. and Range Experiment Station, Forest Service. U.S. Department of Agricul- Thomas, J.W.; Andenon, R.G.; Maser, C.; [and othen]. 1979. Snags. In: mre; 4 p. Thomas, J.W.. tech. ed. Wildrife habitat in managed farens-+he Blue Noon, Barry R. 1981. Techniques for sampling avian habitats. h: Capen. Mountains of Oregon and Washington. Agric. Handb. 553. Washington, David E., ed. The use of multivariate statistics in studies of wildlie habitat. DC: U.S. Department of Agriculture; 60.77. Gen. Tech. Rep. RM-87. Forr Collins. CO: Rocky Mountain Forest and U.S. Deparvnentof AgriculU1re,ForestSewicece1974. Wood handbook: wood Range Experiment Station. Forest Service. U.S. Department of Agriculture; as an engineering material. Agric. Handb. 72. Washington, DC: U.S. De- 42-52. panment of Agriculture; 270 p. Norusis. M.1 1986. SPSS/PC+ for the IBM PC/XT/AT. Chicagq E SPSS U.S. Depanment of Agricul~ure,Forest Sewicc. 1976. Six Rivers National Inc. 324 p. Forest--timber type maps. In-house document an fie at Six Rivcrs Na- Old-GmwrhDcfinitionTaskGroup. 1986. Interim definitiansforold-growlh tional Forest, Eureka. CA. Douglas-fir and mixed.coniferforeSsin the Pacific Northwest and Cali- U.S. Depanmcnt of Agriculture, Forest Service. 1986. Guidelines for select- fornia. Gcn. Tech. Rep. PNW-447. Podand, OR: PacificNonhwcrt Forest ing live or dcad standing tref wildlife habitat. Podand, OR: Pacific and Rangc Experiment Station, Forest Service, U.S. Department of Agricul- Northwest Region; 6 p. ture; 7 p.

USDA Forest SnviceRes. Paper PSW-196. 1989. The Forest Service. U. S. Department of Agriculture, is responsible for Federal leadership in forestry. It carries out this role through four main activities: e Protection and management of resources on 191 million acres of National Forest System lands e Cooperation with State and local governments, forest industries, and private landowners to help protect and manage non-Federal forest and associated range and watershed lands e Participation with other agencies in human resource and community assistance programs to improve living conditions in rural areas e Research on all aspects of Foresuy, rangeland management, and forest resources utilization.

The Pacific Southwest Forest and Range Experiment Station e Represents the research branch of the Forest Service in California, Hawaii, American Samoa and the western Pacific.

Persons of any race, color, national origin, sex, age, religion, or with any handicapping conditions arc welcome lo use and enjoy all facilities. programs, and setvices of theU.S. Depamentof Agricullure. Discrimmation in any fom is strictly against agency policy, and should be repolred to the Secretary of Agriculture. Washington. DC 20250.