UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Food limitation and habitat preference of northern flying squirrels and red squirrels Ransome, Douglas Bruce 1994

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

Item Metadata

Download

Media
831-ubc_1994-0236.pdf [ 1.71MB ]
Metadata
JSON: 831-1.0075246.json
JSON-LD: 831-1.0075246-ld.json
RDF/XML (Pretty): 831-1.0075246-rdf.xml
RDF/JSON: 831-1.0075246-rdf.json
Turtle: 831-1.0075246-turtle.txt
N-Triples: 831-1.0075246-rdf-ntriples.txt
Original Record: 831-1.0075246-source.json
Full Text
831-1.0075246-fulltext.txt
Citation
831-1.0075246.ris

Full Text

FOOD LIMITATION AND HABITAT PREFERENCE OF NORTHERN FLYINGSQUIRRELS AND RED SQUIRRELSbyDouglas Bruce RansomeB.Sc., The University of Windsor, 1986B.Sc., The University of Guelph, 1987A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THEREQUIREMENTS FORTHE DEGREE OF MASTER OFSCIENCEinTHE FACULTY OF GRADUATE STUDIES(Department of Forest Sciences)We accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISHCOLUMBIAApril, 1994© Douglas B. Ransome, 1994In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that theLibrary shaH make itfreely available for reference andstudy. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or herrepresentatives. It is understoodthat copying orpublication of this thesis for financial gain shall not be allowedwithout my writtenpermission.Department of ---The University of British ColumbiaVancouver, CanadaDate —- 9 ‘-.-‘DE-6 (2/88)11ABSTRACTFood limitation and habitat preference of northern flying squirrels (Glaucomyssabrinus) and red squirrels (Tamiasciurus hudsonicus) were examined in the Montane Spruce(MSdm2)and Engelmann-Spruce-Subalpine-Fir (ESSFd2)(transition area) biogeoclimaticzones (Meidinger and Pojar 1991) in the south-central interior of British Columbia. I testedthe hypotheses that: 1) flying squirrel and red squirrel populations are limited by foodavailability, 2) flying squirrel abundance is positively related to the abundance of cavities fornesting and, 3) second-growth stands are sub-optimal habitat for flying squirrels and redsquirrels. Populations were monitored on 6 study areas; 2 second-growth lodgepole pinestands (controls), 2 second-growth stands with food supplementation (treatments: 25-ha blockswith sunflower seeds aerially applied for 3 summers, starting in June 1991), and 2 old-growthlodgepole pine stands. Northern flying squirrel and red squirrel populations were examinedfor two summers, starting June in 1992.From hypothesis 1, I predicted that population size, proportion of adults breeding,body weight, recruitment, and survival would be higher in treatment stands that receivedsupplemental feeding than in control stands. Treatment stands had significantly more flyingsquirrels than control stands (P < 0.00 1). Average densities of flying squirrels were twice ashigh in treatment stands (1.38 and 1.50 squirrels/ha) than control stands (0.64 and 0.68squirrels/ha). In 1992, the abundance of red squirrels was not significantly different betweencontrol and treatment stands (P = 0.74). In 1993, treatment stands had significantly more redsquirrels (P = 0.008) than control stands. The average weight of adult males, survival rates,and the proportions of female and male flying squirrels, and male red squirrels in breedingcondition were not significantly different in control and treatment stands. The proportion offemale red squirrels in breeding condition was significantly higher in control than treatmentstands. Consequently, population size of flying squirrels and red squirrels appeared to be111limited by food availability, but individual squirrels did not appear to benefit from foodsupplementation.An intensive survey of cavities, coupled with subsequent cavity checks, indicated thatflying squirrels did not require cavities and their population size was not limited by theavailability of cavities. These results do not support the hypothesis that flying squirrelabundance is positively related to the abundance of cavities.Population size of flying squirrels, body weight, recruitment, and survival of flyingsquirrels and red squirrels were not significantly different in second-growth and old-growthstands. Population sizes for red squirrels were significantly higher in second-growth than oldgrowth stands. There was no consistent difference in the proportion of squirrels in breedingcondition between second-growth and old-growth stands. These results do not support thehypothesis that second-growth stands are sub-optimal habitat for northern flying squirrels andred squirrels.ivTABLE OF CONTENTSABSTRACTiiLIST OF TABLESviLIST OF FIGURESviiACKNOWLEDGMENTSviiiCHAPTER 1.GENERAL INTRODUCTION1STUDY ANIMALS2OBJECTIVES AND ORGANIZATIONOF THE THESIS4CHAPTER 2. RESPONSE OF NORTHERN FLYING SQUIRRELS AND REDSQUIRRELS TO PROVISION OF SUPPLEMENTAL FOODINTRODUCTION5METHODSStudy area7Experimental design, seed distribution,and cavity surveys7Squirrel population dynamics8Statistical analysis11RESULTSTrappability12Population density14Recruitment15Body weight18Survival rates18Breeding21Movement23Cavity abundance and occupancy23DISCUSSION25VCHAPTER 3. POPULATION DYNAMICS OF NORTHERN FLYING SQUIRRELS ANDRED SQUIRRELS IN SECOND- AND OLD-GROWTH LODGEPOLE PINESTANDS.INTRODUCTION33METHODS34RESULTSTrappability35Population density37Recruitment38Body weight41Survival rates41Breeding44Movement46DISCUSSION48CHAPTER 4:Summary52LITERATURE CITED56viLIST OF TABLESTable 2.1. Estimates of Jolly trappabiity fornorthern flying squirrels and red squirrelsin control and treatment (food supplemented) stands (9 samples/mean). Valuesare mean percent trappability (average (number caught/Jolly-Seber populationestimate) x 100%, Krebs and Boonstra 1984) with 95% confidence intervals inparentheses13Table 2.2. Percentage of adult northern flyingsquirrels and red squirrels in breedingcondition in control and treatment (food supplemented) stands for 1992and 1993 (sample size in parentheses)22Table 2.3. Movement and mean maximum distance moved (MMDM) between consecutivecaptures for flying squirrels and red squirrels(MMDM = mean distance moved+ 3 standard deviations)24Table 3.1. Estimates of Jolly trappability for northern flying squirrels and red squirrels insecond-growth and old-growth stands (9 samples/mean; old-growth - 1993, 7samples/mean). Values are mean percent trappability (average (numbercaught/Jolly-Seber population estimate) x 100%, Krebs and Boonstra 1984)with 95% confidence intervals in parentheses36Table 3.2. Percentage of adult northern flying squirrels and red squirrels in breedingcondition in second-growth and old-growth stands for 1992 and 1993(sample size in parentheses)45Table 3.3. Movement and mean maximum distance moved (MMDM) between consecutivecaptures for flying squirrels and red squirrels (MMDM = mean distance moved+ 3 standard deviations)47viiLIST OF FIGURESFigure 2.1. Estimated population sizes of flying squirrels for control and treatment stands.Population estimates based on all individualscaptured and resident individuals captured are shown. Verticallines indicate time of supplemental food applications16Figure 2.2. Estimated population sizes of red squirrels for control andtreatment stands. Population estimates based on all individualscaptured and resident individuals captured are shown. Verticallines indicate time of supplemental food applications17Figure 2.3. Body weight of adult male flying squirrels (A) andred squirrels (B).Values represent average weight with 95 % confidence intervalsfor control (Ci and C2) and treatment (Ti and T2) stands. Samplesize is recorded above the upper confidence interval19Figure 2.4. Flying squirrel (A) and red squirrel (B) survival rates for control(Cl and C2)and treatment stands (Ti and T2). Survival rates, basedon 28-day intervals, were calculated for four periods: summer 1992,fall, over-winter, and spring 1993. Values are 28-day survival rateswith 95% confidence intervals20Figure 3.1. Estimated population sizes of flying squirrels forsecond- andold-growth stands. Population estimates based on all individualscaptured and resident individuals captured are shown39Figure 3.2. Estimated population sizes of red squirrels for second- andold-growth stands. Population estimates based on all individualscaptured and resident individuals captured are shown40Figure 3.3. Body weight of adult male flying squirrels (A) andred squirrels (B).Values represent average weight with 95 % confidence intervalsfor second-growth (SG1 and SG2) and old-growth (OG1 and 0G2)stands. Sample size is recorded above the upper confidenceinterval42Figure 3.4. Flying squirrel (A) and red squirrel (B) survival rates for second-growth(SG) and old-growth (OG) stands. Survival rates, based on 28-dayintervals, were calculated for four periods: summer 1992, fall,over-winter, and spring 1993. Values are 28-day survival rateswith 95% confidence intervals43viiiACKNOWLEDGMENTSI would like to express my sincere appreciation to Dr. Tom Sullivan, mysupervisor,for his guidance, encouragement, financial assistance, and his constant optimism throughoutmy studies. Special thanks to DonPurdy, Ministry of Forests, Vernon,for his help,materials, and assistance with sunflower seeding. This research was funded through anNSERC operating grant to Dr. Sullivan, a VanDusen Graduate Fellowship to the author, andadditional support from the B. C. Ministry of Forests, Silviculture Section, Kamloops Region.Many thanks to the numerous fieldassistants which helped with the collection of the data, withspecial thanks to Signy Fredricksonand Claudia Ripley. The Bolton family supplied excellentaccommodations which contributedgreatly to the enjoyment of the fieldresearch.Many thanks to my colleagues B.Runciman, C. von Trebra, M. Burwash, B. Booth,and V. Craig for many stimulating conversations and input which helpedcontribute to thesuccessful completion of this research. Special thanks to Lisa Zabek forher help, support, andfor reviewing earlier drafts.I would like to thank my committeemembers, Dr. Peter Marshall, Dr. Walt Kienner,Dr. Kim Cheng, and Dr. John McLean for supplying constructive criticism on earlier drafts.Most importantly, I owe special thanks to my parents for their patience,encouragement, andconstant support throughout my University career.1CHAPTER 1GENERAL INTRODUCTIONCertain processes limit or regulate population numbers. It is accepted that animalpopulations do not continue to increase unabated. Individuals and populations in stabletemperate environments are ultimately limited by the abundance of essential resources, ofwhich food appears to be the most important resource (Lack 1954, Watson 1970, White 1978,Boutin 1990). However, there is some controversy concerning the relative importance of foodcompared with other essential resources or regulating mechanisms inlimiting population size(for example: competition, self-regulating or behavioural mechanisms, predators, or disease)(Watson and Moss 1970, Sinclair 1988).In a recent review of food supplementation experiments with terrestrial vertebrates,Boutin (1990) concluded that a typical response on food-supplemented grids was a 2 to 3-foldincrease in density and enhanced population dynamics. However, evenin the presence ofpresumably unlimited food, populations did not continue to increase.Boutin (1990) suggestedthat other essential resources or regulating mechanisms may have prevented further increasesafter a doubling of density, or that limitations in the experimental designs could haveprevented further increases (method of food application, size of studyarea, and duration ofstudy). To more accurately investigate the influence of food supplementation on populationdynamics, these limitations must be reduced or eliminated.Arboreal sciurids (Tamiasciurus spp.) have been the focus of several recentexperiments testing food limitation and population regulating mechanisms (Sullivan andSullivan 1982, Sullivan 1990, Klenner and Krebs 1991, Sullivan and Kienner 1993). Mystudy examines food limitation and habitat preference of two arboreal sciurids: northern flyingsquirrels (Glaucomys sabrinus Shaw) and red squirrels (Tamiasciurus hudsonicus Erxleben).2STUDY ANIMALSNorthernflying squirrelsNorthern flying squirrels are found in forested regions over most of North America(Wells-Gosling and Heaney 1984). The northern flying squirrel is a small arboreal sciurid thatexhibits a biphasic nocturnal activity pattern (Weigl and Osgood 1974, Mowrey and Zasada1982). They are typically found in habitats dominated by conifers or a mixed coniferous -deciduous overstory (Weigi 1978, Wells-Gosling and Heaney 1984). Stomach and fecalanalyses indicate that flying squirrels consume lichens during the winter and hypogeous fungiduring the summer (McKeever 1960, Maser et al. 1978, 1985, Hall 1991).Flying squirrels use multiple den sites (cavities and constructed nests) on consecutivedays (1-13 nest sites (Weigi and Osgood 1974); 1-7 (Carey 1991)) and often share nests(Maser et al. 1981, Mowrey and Zasada 1982). Flying squirrels are apparently moreabundant, although not consistently (Anthony et al. 1987, Rosenberg and Anthony 1992), inold-growth than second-growth stands (Volts 1986, Carey 1989, Carey et al. 1991a, Carey,eta!. 1992, Witt 1992). Flying squirrel abundance has been positively correlated with thenumber of large live trees and large-diameter ( 76-cm d.b.h.) snags (Volts 1986), althoughnot consistently (Rosenberg and Anthony 1992). Large trees and large-diameter snags aremore abundant in old-growth than second-growth forests (Carey et al. 1991b). The last threeobservations form the primary support for the hypotheses that: i) the abundance of cavitiesarethe primary limiting resource for flying squirrels, and ii) flying squirrels prefer old-growthforests. Food and predation have been suggested as a potential limiting resource andregulating mechanism, respectively, but are considered to be of less importance (Carey 1991,Carey, et a!. 1992).No studies, to date, have critically examined population regulation, limiting factors, orpopulation dynamics of northern flying squirrels.3Red squirrelsRed squirrels are widely distributed throughout the boreal and temperate forests ofNorth America (C. Smith 1981). Red squirrels are a diurnal sciurid that feeds primarily onconifer cones during the winter and fungi, reproductive parts of angiosperms, and vasculartissue of lodgepole pine trees (Pinus contorta) throughout the summer (C. Smith 1968, Kempand Keith 1970, Gurnell 1983). Previous studies have indicated a strong relationship betweenthe abundance of conifer cones and red squirrel densities (C. Smith 1968, M. Smith 1968,Kemp and Keith 1970, Gurnell 1983). Red squirrels exhibit two modes of food hoarding;larder-hoarding (storage of cones in a few highly concentrated sites - midden&) is morecommon in western North America, while scatter-hoarding (storage of one or a few cones innumerous caches) is more common in eastern North America (C. Smith 1968, Gurnell 1984,Hurly and Roberston 1987, Dempsey and Keppie 1993). Middens are often located near thecenter of well advertised and strongly defended territories. The interrelationship of territorialbehaviour, food abundance, and red squirrel densities has recently been examined (Kiennerand Krebs 1991, Klenner 1991). During periods when red squirrel densities are low,populations appear to be limited by food availability. However, when population densities arehigh, the abundance of red squirrels is regulated by territorial behaviour, despite the presenceof excess food (Kienner and Krebs 1991, Kienner 1991).4OBJECTIVES AND ORGANIZATION OF THE THESISThe two main objectives of this thesis are: 1) to examine population dynamics andfood limitation of northern flying squirrels and red squirrels (Chapter 2) and 2) to determine ifsecond-growth forests are sub-optimal habitat for flying squirrels and red squirrels (Chapter 3).In Chapter 2, I examine the effect of large-scale food supplementation on northernflying squirrel and red squirrel population dynamics. I test the hypothesis that flying squirreland red squirrel populations are limited by the abundance of food. In addition, I evaluatecavity availability as a potential limiting resource for flying squirrel populations.In Chapter 3, I compare characteristics of flying squirrel and red squirrel populations insecond- and old-growth lodgepole pine forests to test the hypothesis that second-growthlodgepole pine forests are sub-optimal habitat for northern flying squirrels and red squirrels.5CHAPTER 2RESPONSE OF NORTHERN FLYING SQUIRRELS AND RED SQUIRRELS TOPROVISION OF SUPPLEMENTAL FOODINTRODUCTIONThe influence of food abundance upon animal population dynamics has been welldocumented (Watson 1970, Boutin 1990). In a recent review of food supplementationexperiments with terrestrial vertebrates, Boutin (1990) reached the following generalizationsabout population dynamics on food-supplemented grids: home ranges were smaller, averageadult and juvenile body weights were heavier, litter size increased, population densityincreased (2-3-fold increase over unfed populations), growth rates were higher, and moreenergy was spent on territorial defense. Results from population studies with arboreal sciuridsare consistent with some of Boutin’s (1990) conclusions, indicating that these species may belimited by food availability. Red squirrels and Douglas squirrels (T. douglasli) haveresponded positively to food supplementation experiments using sunflower seeds (Helianthusannuus). Increased densities of 50-600% over unfed populations have been reported (Sullivanand Sullivan 1982, Sullivan 1990, Kienner and Kiebs 1991, Sullivan and Kienner 1993) whilelonger breeding seasons, second litters, and smaller home ranges have been observed (Sullivanand Sullivan 1982, Klenner and Krebs 1991). However, supplemental feeding had little effecton adult survival and body weights (Klenner and Krebs 1991, Sullivan and Kienner 1993),juvenile survival and growth rates (Sullivan and Sullivan 1982, Sullivan 1990, Klenner andKrebs 1991), and proportion of adult squirrels breeding (Sullivan 1990, Klenner and Krebs1991, Sullivan and Kienner 1993). In contrast, in a study of fox squirrel (Sciuru.c niger) andgray squirrel (S. carolinensis) responses to late spring and early summer food shortages,6Koprowski (1991) concluded that food shortage, which influences juvenile survival and bodycondition, can be critical for populations of Sciurus.The lack of a consistent pattern in population responses of arboreal squirrels tosupplemental food may have resulted from limitations in experimental designs. Supplementalfood has often been applied in highly concentrated, thus defensible, sources (5-litre pails orpiles). After an initial influx of transients (Boutin 1990, Sullivan 1990) not all individualswould have access to the food, thus, it may become limited. Secondly, the period whensupplemental food has been available may have been too brief to affect the animals (Boutin1990, Kienner and Krebs 1991). Finally, the relative scale at which supplemental food hasbeen applied (9 ha, Sullivan 1990; 13 ha, Klenner and Krebs 1991; 15 ha, Sullivan andKienner 1993), compared to the corresponding live-trapping grid (9 ha), has been small.Consequently, the influence of immigration may have been large enough to mask residentsquirrel population responses to supplemental food (Boutin 1990, Koford 1992).In this study, the influence of food on flying squirrel and red squirrel populationdynamics was more intensively investigated than previous studies by applying sunflower seedsaerially, in a uniform distribution, over 25-ha treatment blocks for 3 years. Recent studies(Volts 1986, Carey 1991, Carey et al. 1992) have indicated that the availability of cavitieswas the primary limiting resource for northern flying squirrels. Therefore, an additionalobjective of this thesis is to determine if cavity availability limits flying squirrel populations.This research specifically addresses the following hypotheses, phrased as predictions:Hi: Food supplementation will induce higher densities of northern flying squirrels and redsquirrels compared to unfed populations.H2: Population dynamics (breeding, body weight, recruitment, and survival rate) of flyingsquirrels and red squirrels will be higher in food supplemented stands than control stands.H3: The abundance of flying squirrels will be positively related to the abundance of cavities.7METHODSStudy areaThe McGregor Creek study area (500 23’ N; 119° 36’ W) was located 60 km west ofVernon in the south-central interior of British Columbia. Site elevation ranged from 1356 mto 1478 m in the Montane Spruce (MS2)and Engelmann Spruce Subalpine Fir (ESSFJC2)(transition area) biogeoclimatic zones (Meidinger and Pojar 1991). The climate ischaracterized by warm, dry summers and cold, dry winters with mean July and Januarytemperatures of + 16 and -10 °c, respectively, and an annual precipitation of 40 cm. Thedominant coniferous species in mature stands are Engelmann (Picea engelmannil) and hybridspruce (P. engelmannhi-P. glauca) and varying amounts of Douglas-fir (Pseudotsugamenziesii) and subalpine fir (Abies lasiocarpa).Approximately 900 ha were logged in 1963 to 1970 and regenerated naturally tolodgepole pine. Approximately 450 ha were thinned over a 6-year period starting in 1977 andwere fertilized (200 kg/ha) with agricultural grade urea ((NH2)Co+ elemental sulphur) in1987. Four thinned and fertilized lodgepole pine stands were selected for the study: 2treatment stands (each 36 ha; age 21 and 28 years) and 2 control stands (17.8 and 38.3 ha;age28 and 20 years, respectively). The average d.b.h. (diameter at breast height, 1.3 m abovethesoil surface) of trees in these stands ranged from 13.8 ± 0.2 cm to 16.8 ± 0.3 cm.Experimental design, seed distribution, and cavity surveysThe four study blocks were selected in the spring of 1991. Selection of study blockswas based primarily on areas of suitable habitat and homogeneity of forest characteristics(dominance of young lodgepole pine trees and similarity of understory characteristics).Sunflower seeds were applied (20 kg/halmonth) to twenty-five-hectare treatment blocks by8helicopter in a uniform distribution from early June to October, 1991 (5 applications), May toOctober, 1992 (6 applications), and June to August, 1993 (3 applications). Timing of initialapplication each year was dependent upon the duration of snow cover. The rate of seedapplication (20 kg/ha) is the standard rate used to deter redsquirrel feeding damage in younglodgepole pine stands (Sullivan and Kienner 1993).In the summer of 1992, a survey of all snags and cavities was conducted in control andtreatment stands. Two to four individuals walked parallel transects, while maintaining visualcontact, and completely surveyed 13.75 ha enclosing each 9-ha live-trapping grid. All snagswere marked, the location noted, and the number of cavities recorded. Cavities wereexamined for suitability as den sites and occupancy. Cavities were classified as suitable densites if they contained nesting material or were considered structurally functional as a den site.A cavity was considered occupied if a flying squirrel exited the cavity after the base of thesnag was struck 2-3 times. In 1993, all cavities were checked for occupancy by flyingsquirrels, 3-4 times during the summer, at 2-week intervalsstarting at the time of lactation.Squirrel population dynamicsEach of the 4 blocks contained a 9-ha live-trapping grid consisting of 96 (6 X 16)stations located at 30-m intervals. Traps were mounted approximately 1.5 m above theground, on a tree trunk, at alternating stations. Trap stationswere staggered with respect toadjacent lines, resulting in 48 traps per live-trapping grid at anequidistance of 42.4 m.Tomahawk live-traps (Model 201, Tomahawk Live Trap Co., Tomahawk, Wisconsin)equipped with a nest box (1-litre plastic jar with coarse brown cotton) and a trap cover (rolledroofing cut and bent to offer protection from wind and rain on 3sides) were used to capturenorthern flying squirrels and red squirrels. Squirrel populations were live-trapped every 2weeks from June to October, 1992 (10 trap sessions) and April 15to August 25, 1993 (10 trapsessions). During the 1992 trap sessions, traps were set on the evening (one hour before dark)9of day 1, checked in the morning and afternoon of day 2, thenlocked open. Traps were reseton the evening of day 2, checked in the morning and afternoon of day 3, then locked openuntil the next trap session. In 1993 the morning and afternoon trap sessions were combined,resulting in one check between 10:30 am and 12:30 pm. Persistence of snow and limitedaccessibility to the stands in the spring of 1993 necessitated the condensed trap schedule. Thetrap schedule was maintained throughout the 1993 trap session for consistency and efficiency.Traps were baited for a one-week prebaiting period each spring, and during the timewhen the traps were set, with a mixture of peanut butterand whole oats. All squirrelscaptured were identified with individually numbered eartags (1 ear tagged for red squirrelsand 2 ears tagged for flying squirrels). When animals werecaptured, ear tag number/s,location, weight (to ±5 g on a Pesola spring balance), sex, and breeding condition wererecorded. Female breeding condition was evaluated by palpation of the mammaries andclassified as ‘small’ (non-breeding) or ‘large’ (breeding).Male breeding condition wasevaluated by palpating the testes and classified as either ‘abdominal’ (non-breeding) or‘scrotal’ (breeding).Parameters measured were trappability, population density, proportion of adultsquirrels in breeding condition, body weight, recruitment,survival, and average distancemoved between consecutive trap nights. Comparisons of population dynamics between controland treatment stands were used to evaluate the effect of large-scale food supplementation onflying squirrel and red squirrel populations. Jolly trappability (Jolly 1965, Jolly and Dickson1983, Krebs and Boonstra 1984) was used as an indication of the proportion of the populationtrapped. Population size was estimated for each trap period using the Jolly-Seber model ofpopulation estimation (Seber 1982). However, the reliability of Jolly-Seber estimatesdecreases when populations sizes are very low and no taggedanimals are captured (Krebs etal. 1986). Consequently, the minimum-number-of-animals-known-to-be-alive (MNA, Krebs1966) was calculated as a precautionary measure, and to indicate the lower limits of the JollySeber estimates. However, all comparisons and statistical testswere based on the Jolly-Seber10estimates for reasons indicated by Jolly and Dickson (1983). In 1992, the last two trappingperiods were eliminated from population estimates due toextensive black bear (Ursusamericanus) disturbance. Resident population size was based on those individuals captured atleast twice (Sullivan and Kienner 1993). Recruits were classified as new squirrels captured atleast twice. Population densities were based on averageestimated population size divided bythe effective trapping area (trapping grid + 1/2 the mean-maximum-distance-moved, MMDM(see Witt 1992)). Since 90% of the observed distance moved between consecutive trap nightslie within 3 standard deviations of the mean (Tchebysheff’s Theorem, Mendenhall 1983, p.47), MMDM was calculated as the mean distance moved between consecutive trap periodsplus three standard deviations. The percentage of sexually mature adults in a series of weightclasses, coupled with the lowest weight attained by anyknown adult squirrel, was used todetermine age categories. Red squirrels and flying squirrels weighing less than 170 ± 5 g and105 ± 5 g, respectively, were never sexually mature and were classified as juveniles. Noknown adults weighed less than these values; adult flying squirrels did not drop below 105 g,while red squirrels did not drop below 175 g. Reproduction was evaluated from the proportionof adult females and adult males in breeding condition.Comparisons of the proportion offemales in breeding condition were based on 1992 and 1993 estimates, while proportion ofbreeding males was based on 1993 estimates only. Earlysnow melt and timing of the first trapsession in 1992 resulted in all male squirrels captured being abdominal. Average proportion ofsquirrels breeding was weighted according to sample sizeand averaged over two replicates.Body weight comparisons between control and treatment stands were based on the averageweight of each resident adult male and averaged for eachstand. Trappability, populationdensity, reproduction, recruitment, survival rates, and movement were calculated by SmallMammal Programs For Mark-Recapture Data Analysis (C.J. Krebs, Department of Zoology,University of British Columbia, Vancouver, B.C. V6T 2A9) for each replicate and year for theflying squirrel and red squirrel population data.11Statistical analysisNonparametric statistical procedures arealmost as capable of detecting differencesamong population parameters as parameteric methods when normality and other assumptionsare satisfied (Mendenhall 1883, p 606;Zar 1984, p. 138). They often are more powerful indetecting population differences whenthe assumptions are not satisfied (Mendenhall 1883, p606). The efficiency of nonparametric procedures, relative to parametricones, is quite highfor small sample sizes; n 10 (Steel and Tone 1980, pg. 533-534). Since the same animalswere often captured in consecutive sampling periods (lack of independence), and sample sizeswere small (2 replicates of 7 trap periods in 1992, and 9 trap sessions in 1993), Wilcoxon’sranked sign test (Sokal and Roif 1981, pp. 448-449) was usedto test for significantdifferences. This test was based onan average population size for controland treatmentstands. A chi-squared test was use toindicate the degrees of difference in the proportion ofadult squirrels breeding. Significantdifferences between the average weight of adult malesquirrels (based on a mean weight foreach squirrel) between control and treatment stands weretested by analysis-of-variance (ANOVA) for 1992 and 1993 separately. Significant differencesin recruitment, based on 1993 data only, between control and treatment stands was tested byan ANOVA. Wilcoxon’s ranked signtests and ANOVAs were carried out using the SPSSstatistical analysis package (SPSS Inc., 444 North Michigan Ave., Chicago,Ii., 60611). In allanalyses, unless otherwise noted, the level of significance was a = 0.05.12RESULTSTrappabilityJolly trappability for female and male northern flying squirrels and red squirrels wereestimated for each stand and year separately (Table 2.1). Overall trappabiity of flyingsquirrels was greater in the control stands than treatment stands (average of 50.1 % acrosscontrols vs. 36.4% for treatments). However, none of the eight pair-wise comparisons weresignificantly different (based on nonoverlapping 95% confidence intervals). In control andtreatment stands 51.1% and 27.6%, respectively, of the flying squirrels were captured onmore than two occasions.Similarly, overall trappability for red squirrels was greater in the control stands(66.4%) than treatment stands (52.7%), however, none of the eight pair-wise comparisonswere significantly different. There was no consistent difference between female and maletrappability.13Table 2.1. Estimates of Joiiy trappability for northern flying squirrels and red squirrels incontrol and treatment (food supplemented) stands (9 samples/mean). Values are meanpercent trappabiity (average (number caught/Jolly-Seber population estimate) x 100%,Krebs and Boonstra 1984) with 95% confidence intervals in parentheses.1992 1993FEMALE MALE FEMALE MALEFLYING SOUIRRELControl1 55.2 66.7 47.7 34.9(30.6-79.9) (49.5-83.8) (30.0-65.3) (12.8-57.0)2 46.2 49.4 47.0 53.5(18.0-74.5) (28.6-70.3) (17.6-76.4) (30.3-76.9)Treatment1 61.7 31.6 33.0 36.6(30.9-92.4) (8.9-54.2) (15.4-50.6) (7.0-66. 1)2 31.0 42.7 24.1 30.1(15.4-46.6) (20.6-64.7) (0.3-48.0) (17.9-43.4)RED SOUIRRELControl1 63.0 66.9 58.7 71.7(44.1-81.9) (53.7-80.1) (36.1-81.2) (55.3-88.0)2 65.7 63.7 64.1 76.7(45.7-85.7) (48.1-79.2) (45.0-83.2) (64. 1-89.3)Treatment1 54.4 41.4 54.0 64.1(25.7-85.7) (10.4-72.5) (25.8-82.2) (39.8-88.4)2 51.2 53.3 54.6 47.8(27.4-75.0) (30. 1-76.6) (29.9-79.3) (21.1-74.4)14Population densityOne-hundred-twenty-one flying squirrels were captured a total of 380 times during twoseasons of live-trapping. Forty-five flying squirrels were captured in control stands and 76flying squirrels were captured in treatment stands. Wilcoxon’s ranked sign test indicated thatthe estimated population size of flying squirrels, based on all squirrels captured (Figure 2.1),in treatment stands was greater than in control stands (P < 0.001). This relationship did notchange when the population estimates were based on resident squirrels (P < 0.001). Thecontrol replicates were not significantly different from one another (P = 0.51), nor were thetreatment replicates (P = 0.72). Similarly, when the population estimates were based onresident squirrels, control replicates were not significantly different (P = 0.91) from eachother, nor were the treatment replicates (P= 0.53).There was a two-fold increase in the average flying squirrel density in treatment stands(1.38 and 1.50 squirrels/ha) than control stands (0.64 and 0.68 squirrels /ha). There was a1.8-fold increase in the average flying squirrel density, based on MNA, in treatment stands(1.00 and 1.05 squirrels/ha) compared to control stands (0.57 and 0.54 squirrels/ha). Resultswere similar when Jolly-Seber densities were based on resident squirrels (controls: 0.60 and0.60 squirrels/ha; treatments: 1.15 and 1.21 squirrels/ha).The abundance of red squirrels (based on all squirrels captured) in treatment stands wassignificantly greater (P = 0.017) than control stands (Figure 2.2). This relationship did notchange when the population estimates were based on resident squirrels (P = 0.015). Thenumbers of red squirrels in control replicates were not significantly different from each other(P = 0.20) while the numbers in treatment replicates were (T2 > Ti, P = 0.03). Populationestimates, based on resident squirrels, in control stands were significantly different from eachother (P = 0.01), as were the treatment stands (Ti > T2, P = 0.015). The average densityof red squirrels in treatment stands (Ti: 2.65, T2: 3.15 squirrels/ha) was 1.8-times larger thanin control stands (Cl: 1.56, C2: 1.71 squirrels/ha). Population density, based on MNA, in15treatment stands (Ti: 2.27, T2: 2.49) was 1.6-times larger than in control stands (Cl: 1.35,C2: 1.55 squirrels/ha).I further examined red squirrel population size in control and treatment stands for eachyear separately. In 1992, the estimated population size of red squirrels in control (Ci: 24.1,C2: 26.3), and treatment stands (Ti: 22.7, T2: 24.8) was not significantly different from eachother (P = 0.74). In 1993, Wilcoxon’s ranked sign test indicated that estimated populationsize was significantly greater in treatment (Ti: 30.3, T2: 37.841) than control stands (Cl:18.5, C2: 20.0) (P = 0.008).RecruitmentThe number of recruits for flying squirrels in control stands (average for 16 trapsessions ± 95% CI; 0.20 ± 0.26 and 0.50 ± 0.79) did not differ from treatment stands (0.60± 0.52 and 0.50 ± 0.79) (ANOVA, P = 0.83). The number of recruits for red squirrels incontrol stands (1.00 ± 1.49 and 1.70 ± 0.97) did not differ from treatment stands (1.80 ±1.75 and 2.00 ± 1.75) (P = 0.36).A. ALL INDIVIDUALS16CONTROL 1 -*- TREATMENT 1•4( CONTROL 2 TREATMENT 2Figure 2.1. Estimated population sizes of flying squirrels for control and treatmentstands. Population estimates based on all individuals captured and residentindividuals captured are shown. Vertical lines indicate time of supplemental foodapplications.-4I.19923 J1993B. RESIDENTS403020100MJ1992J A SO AM J J A1993CONTROL 1 * TREATMENT 14_ CONTROL 2 9 TREATMENT 217.tIzCCIA. ALL INDWIDUALS100Figure 2.2. Estimated population sizes of red squirrels for control and treatmentstands. Population estimates based on all individuals captured and residentindividuals captured are shown. Vertical lines indicate time of supplemental foodapplications.1992M J J A SO AM J J A1993CONTROL 1 * TREATMENT 1CONTROL 2 9 TREATMENT 2B. RESIDENTS6050403020M J J A SO AM J J A1992 1993— CONTROL 1 * TREATMENT 1CONTROL 2 TREATMENT 218Body weightThere was no significant difference in the average weight of adult male flying squirrelbetween control and treatment stands (Figure 2.3A), in 1992 (ANOVA, P = 0.86) and 1993(ANOVA, P = 0.88). Average weights in control and treatment stands were 140.9 g and141.8 g, respectively.In 1992, the average weight of red squirrels (Figure 2.3B) in control stands (mean ±95% CI; Cl: 237.4 ± 11.2 g, C2: 229.3 ± 12.1 g) was not significantly different fromtreatment stands (Ti: 238.7 ± 8.4 g, T2: 202.5 ± 11.6 g) (ANOVA, P = 0.19). In 1993,the weight of red squirrels in treatment stands (Ti: 239.0 ± 9.5 g, T2: 246.9 ± 6.6 g),although heavier, were not significantly different from control stands (Cl: 229.0 ± 9.4 g, C2:235.3 ± 8.8 g) (ANOVA, P = 0.28).Survival ratesJolly-survival rates, based on four-week intervals, were calculated for 4 periods todetermine if supplemental feeding enhanced summer, late fall, over winter, and springsurvival. Supplemental feeding had little influence on survival during all 4 periods fornorthern flying squirrels (Figure 2.4A) and red squirrels (Figure 2.4B), based on overlapping95% confidence intervals.19A) NORTHERN FLYING SQUIRRELSCz155150145140135130Ci C2 Ti T21992B) RED SQUIRRELS260250240230220210200190180Ci C21992Ti T2Ci C2 Ti T21993Ci C2 Ti T21993Figure 2.3. Body weight of adult male flying squirrels (A) and red squirrels (B).Values represent average weight with 95% confidence intervals for control (Cland C2) and treatment (Ti and T2) stands. Sample sizes are recorded above theupper confidence interval.3IGIiI±zzzzI I I I I I I I I I I I I IU20•f•H•4•ih*Ci C2 TI T2 Ci C2 Ti T2 Ci C2 Ti T2 CI C2 TI T2Figure 2.4. Flying squirrel (A) and red squirrel (B) survival rates for control (Cland C2)and treatment stands (Ti and T2). Survival rates, based on 28-dayintervals, were calculated for four periods: summer 1992, faIl, over-winter, andspring 1993. Values are 28-day survival rates with 95% confidence intervals.A) NORTHERN FLYING SQUIRRELS21.5C,,B) RED SQUIRRELS1.41.2i0.80.60.40.20SUMMER FALLCiC2T1T2 C1C2T1T2 C1C2T1T2 C1C2T1T2WINTER1992 1992-93SPRING199321BreedingThe percentages of adult squirrels in breeding condition in control and treatment standsare shown in Table 2.2. Supplemental feeding had little influence on the proportion of female(x2 = 3.54, df= 3,0.25< P < 0.50)andmale(x2= 0.92, df= 1, 0.25 <P <0.50)flying squirrels in breeding condition. The weighted-average proportion of females and malesin breeding condition across all stands and years was 38.8% and 59.6%, respectively.Eleven and nine juvenile flying squirrels were captured in all stands in 1992 and 1993,respectively. Two juveniles first appeared in Treatment 2 during the week of June 27, 1992with the majority captured in July. In 1993, 2 juveniles were captured in late July inTreatment 1 with the remainder appearing in August.Food supplemented stands had a lower proportion of female red squirrels breeding(1992: Ti - 0%, T2 - 2.8%; 1993: Ti - 28.9%, T2 - 33.3%) than in control stands (1992: Ci- 5.7%, C2 - 12.8%; 1993: Ci - 46.8%, C2 - 28.0%) (x2 = 14.0, df = 3, P < 0.005). Theproportions of males breeding in control stands (Cl - 82.8%, C2 - 91.3%) were notsignificantly different from treatment stands (Ti - 87.5%, T2 - 93.3%) (x2 = 2.26, df = 1,0.10 <P < 0.25).In 1992, young-of-the-year (30 in total) red squirrels appeared after June 13 with themajority appearing in the last week of June and throughout July. Fewer juveniles werecaptured (6) in 1993, with all appearing in mid- to late August.22Table 2.2. Percentage of adult northern flying squirrels and red squirrels in breedingcondition in control and treatment (food supplemented) stands for 1992 and 1993 (sample sizein parentheses).1992 1993FEMALE MALE FEMALE MALEFLYING SQUIRRELControl1 26.3 0 39.1 63.6(19) (30) (23) (11)2 37.5 0 43.8 63.0(16) (92) (16) (27)Treatment1 22.2 0 55.6 57.1(18) (25) (27) (28)2 34.8 0 50.0 57.6(23) (25) (10) (33)RED SqUIRREL1Controli 57a 0 468a 82.8(35) (58) (47) (64)2 128a 0 280a 91.3(86) (52) (50) (80)Treatment1 0b 0 289b 87.5(50) (36) (90) (80)2 28b 0 333b 93.2(72) (26) (108) (74)1) Proportion of female red squirrels breeding in control stands (a) was significantly higher than in treatment stands (b)(x2 = 14.0, df = 3, P < 0.005)23MovementMean maximum distance moved (MMDM) between consecutive trap periods are shownin Table 2.3. The smallest and largest MMDM traveled both for flying squirrels and redsquirrels occurred in Cl. The low sample size reflects the low proportion of flying squirrelscaught in two successive trap periods. Average MMDM in control and treatment stands was93.6 m and 78.3 m, respectively.The average distance traveled by red squirrels between consecutive trap periods incontrol and treatment stands are shown in Table 2.3. There was a two-fold increase in theaverage MMDM for red squirrels in control stands (71.8 m) compared to treatment stands(35.3 m). Average distance traveled between consecutive trap periods was twice as great in1993 than 1992, in all stands.Cavity abundance and occupancyIn 1992, control and treatment stands were surveyed to determine the abundance ofsnags with cavities. Control 1 had six cavities, of which five were deemed functional.Control 2 no cavities. Treatment 2 had four functional cavities, of which three were located inthe same snag. Treatment 1 contained one functional cavity. Of a total of 10 cavities foundon four 13-ha areas, only one was occupied by a flying squirrel during 45 cavity checks (10cavities checked 4 - 5 times biweekly). All other cavities were unoccupied.24Table 2.3. Movement and mean maximum distance moved (MMDM) between consecutive capturesfor flying squirrels and red squirrels (MMDM = mean distance moved + 3 standard deviations).number of mean distance standard MMDMobservations moved (m) deviation (m)FLYING SOUIRRELSControl 11992 2 111.1 4.8 125.51993 8 446 2.5 52.1Control 21993 6 77.8 8.5 103.3Treatment 11992 3 56.3 3.6 67.01993 4 73.0 5.1 88.3Treatment 21992 3 68.3 3.8 79.7RED SOUIRRELSControl 11992 27 40.1 6.0 58.21993 28 104.2 9.1 131.6Control 21992 61 42.3 5.8 59.71993 30 100.4 8.4 125.6Treatment 11992 17 17.0 4.9 21.71993 33 59.5 7.6 82.3Treatment 21992 19 16.6 4.8 31.01993 48 48.2 6.9 68.925DISCUSSIONNorthern flying squirrel and red squirrel populations were trapped for two years toexamine the influence of large-scale supplemental feeding on population dynamics. Trappingof flying squirrel and red squirrel populations was initiated during the second year of foodsupplementation, and was terminated at the end of the 1993 trapping season. Therefore, thisstudy lacked pretreatment and post treatment data on squirrel populations. Consequently, thefollowing conclusions are based on the assumption that initial flying squirrel and red squirrelpopulation dynamics were similar in control and treatment stands prior to foodsupplementation. Future studies would strengthen their conclusions by incorporatingpretreatment data or increasing the number of replicates in their experimental design.However, population dynamics (breeding, density, weight, recruitment, survival rates, andMMDM) of flying squirrels and red squirrels, in general, were similar within control andtreatment replicates. Therefore, despite the lack of pretreatment data and the low number ofreplicates, I am confident that the population parameters measured for flying squirrels and redsquirrels accurately reflect the influence of large-scale supplemental feeding on thesepopulations. It should be noted that the Wilcoxon’s ranked sign tests were based on populationestimates repeated over time for the same stand, and may be construed as temporallypseudoreplicated (Hurlbert 1984).Aerial application of sunflower seeds (20 kg/ha/month) in a uniform distribution over25-ha treatment blocks was conducted for three years. Sunflower seeds, in seed plots, werecompletely consumed or removed to cache sites prior to the next application (Kohier 1993).Therefore, this application rate may not have been high enough to completely eliminate foodlimitation in treatment stands. Since the sunflower seeds were consumed or removed after twoweeks, future studies should double the application rate.26This research addressed three hypotheses, the first of which was:Hi: Food supplementation will induce higher densities of northern flying squirrels and redsquirrels than unfed populations.Northern flying squirrels were live-trapped in two controls and two large-scale foodsupplemented stands for two years. In response to supplemental food, the density of flyingsquirrels was twice as high in treatment stands (1.38 and 1.50 squirrels/ha) than control stands(0.64 and 0.68 squirrels/ha). These results strongly indicate that the abundance of flyingsquirrels in young second-growth lodgepole pine stands were limited by food availability.The densities of flying squirrels in lodgepole pine stands are comparable to densitiesfound in other forest types. Rosenberg and Anthony (1991) reported the highest densities: 2.0squirrels/ha in second-growth, and 2.3 squirrels/ha in old-growth stands. Flying squirreldensities in northern Washington averaged 0.2 squirrels/ha in second-growth and 0.5squirrels/ha in old-growth stands (Carey et at. 1992). Intermediate densities of flying squirrelswere reported for western Oregon: second-growth - 0.12 squirrels/ha, old-growth - 0.85squirrels/ha (Witt 1992); second-growth - 0.9 squirrels/ha, old-growth stands- 2.0 squirrels/ha(Carey et at. 1992).Large-scale food supplementation appeared to have little influence on red squirrelpopulation sizes, based on all squirrels captured, during the first year of live trapping (secondyear of food supplementation). However, in 1993, population sizes of red squirrels intreatment stands were significantly greater than those in control stands.Rusch and Reeder (1978) suggested that territorial behavior regulated densities of redsquirrels about a mean determined by food availability in years of poor cone crops. Similarly,Kienner and Krebs (1991) reported that red squirrel densities increased during foodsupplementation experiments. However, the presence of excess food indicated that factorsother than food availability limited the population at high densities. The abundance of natural27food in control and treatment stands, in 1992, may have been high enough to minimize theinfluence of food supplementation. Alternatively, food supplementation may havesignificantly increased densities of red squirrels in both years, but other factors may havemasked this influence in 1992.If natural food was over-abundant in 1992, to the extent that territorial behaviour wasthe primary population regulating mechanism, then further increases in population sizeresulting from food supplementation would not occur (Rusch and Reeder 1978, Klenner 1991).In 1993, the average population size of red squirrels in treatment stands (Ti: 30.6, T2: 37.8squirrels/stand) exceeded those for control and treatment stands in 1992 (Cl: 24.2, C2: 26.3,Ti: 22.7, T2: 24.8). Consequently, food supplemented stands had higher population sizes in1993 than 1992. This indicated that population sizes in 1992 were below the level in whichfood was no longer a limiting resource. In addition, there was a two-fold increase in theaverage distance traveled in control stands compared to treatment stands in 1992 and 1993. Ifdistance traveled between consecutive trap periods reflects distance traveled while foraging,then red squirrels foraged over twice the area in control than treatment stands. This wouldindicate a greater availability of food in treatment than control stands in 1992 and 1993.I suggest that red squirrels were limited by food availability in both years of the study.Due to the excessive food present in the treatment stands, and short periods when food wasavailable in the traps (4 nights/mo.), initial capture rates were lower in treatment stands. Theconstant increase in population size in treatment stands in 1992 (Figure 2.2) may moreaccurately reflect a slow learning response of red squirrels to traps and trap-food than aconstant increase in population size. In contrast, in 1992, initial captures of red squirrels incontrol stands occurred rapidly as a consequence of larger foraging routes and greaterdependence on trap food (as reflected in greater trappability).Despite the slow response of red squirrels in treatment stands, food supplementationappeared to have a significant influence on flying squirrel and red squirrel densities. These28results support the hypothesis that flying squirrel and red squirrel populations were limited byfood availability during this study.H2: Population dynamics (reproduction, body weight, recruitment, and survival rates) ofnorthern flying squirrels and red squirrels will be higher in food supplemented standsthan control stands.Supplemental food had little influence on reproduction, adult body weight, recruitment,and adult survival of northern flying squirrels and red squirrels. These results are consistentwith other food supplementation studies with arboreal sciurids. These studies found thatsupplemental feeding had little effect on adult survival and body weights (Kienner and Krebs1991, Sullivan and Kienner 1993) and proportion of adult squirrels breeding (Sullivan 1990,Kienner and Krebs 1991, Sullivan and Kienner 1993). As in my study, these studies showed asignificant increase in red squirrel densities in response to food supplementation. Boutin(1990) found that in response to supplemental food, terrestrial vertebrates, in general, showedincreased reproduction, higher body weights, and improved survival. Kienner and Krebs(1991) suggested that despite the lack of an increase in reproduction, body weight, andsurvival, food availability may influence these parameters in red squirrels. They suggestedthat the duration of food supplementation may have been too brief to affect these animals.Alternatively, the method and scale of food supplementation may have prevented detection ofchanges in these parameters. In the present study I adjusted the experimental design tocompensate for these limitations. The results remained the same: reproduction, adult bodyweight, recruitment, and adult survival appeared to be unaffected by food availability.The proportion of flying squirrels in breeding condition was not influenced bysupplemental feeding. The depletion of supplemental food prior to the next seeding wouldindicate that food still may have been limited in treatment stands. Although the density offlying squirrels increased in response to supplemental feeding, the food availability per squirrel29could have remained the same. Consequently, the poor influence of food supplementation onflying squirrel breeding may have resulted from food limitation.A poor relationship between current food availability and proportion of female redsquirrels in breeding condition may be expected. Kemp and Keith (1970) noted that anincrease in female reproduction occurred, before a good cone crop, even though the cone cropfor the previous two years was poor. They suggested that female reproduction was lessdependent upon current food availability and more dependent upon anticipated coneproduction. In a test of this hypothesis, Larsen (1990) artificially supplemented (double theabundance of cones), removed (approx. 30%), and left unaltered the number of cones infemale red squirrel middens. He found no significant difference in the proportion of femalesbearing offspring among the three groups the following spring. Litter sizes were not affectedby cone alteration; however, the number of females that had young surviving to the fall wassignificantly greater for cone-added females than control and cone-depleted females.In this study, large-scale food supplementation for three years failed to enhance theproportion of breeding female red squirrels. These results would indicate that fluctuations inthe proportion of female red squirrels in breeding condition may be better explained by ananticipatory response to future natural food availability than current food availability.Northern flying squirrels may have been limited by food availability during this study.The fact that body weights and survival rates were not significantly different in control andtreatment stands may reflect this food limitation. Alternatively, Thorington and Heaney(1981) examined body mass, wing loading, and glide ratios of flying squirrels. Since thesurface-area to volume ratio increases at a rate of 2:3, larger squirrels have a higher wingloading (each unit of wing membrane must support a larger weight). For heavy flyingsquirrels to maintain similar glide ratios (horizontal distance/vertical drop) as light squirrels,they must glide faster. Thorington and Heany (1981) suggested that the most lightly loadedsquirrels were best adapted to gliding in forested areas where slower flight speeds and highmaneuverability were important. Therefore, the poor influence of supplemental feeding on30flying squirrel body weights (and concomitant survival rates) may reflect the optimization ofbody weight with gliding efficiency.Wauters and Dhondt (1989) found that survival was positively correlated with bodyweight in red squirrels (Sciurus vulgaris) in Belgium. In addition, large squirrels displacedsmall squirrels from higher quality habitats, which further enhanced their survival. C. Smith(1968) showed that locigepole pine cones yielded their energy, for red squirrels, at 1/4 the rateof Douglas-fir cones. Consequently, lodgepole pine stands have been considered the leastpreferred coniferous habitat for red squirrels (C. Smith 1968, Gurnell 1984). Since redsquirrels may benefit from larger body weights, and lodgepole pine stands supply a lowerquality food source, I expected a significant increase in adult body weights and survival inresponse to supplemental feeding. I offer two potential reasons to explain the lack of thatresponse: 1) due to an increase in intraspecific and interspecific competition for thesupplemental food, food was still limiting, or 2) flying squirrels and red squirrels were at their‘optimum weight’ and further increases in weight would be a disadvantage.Red squirrels may have been limited by food availability during this study. However,Kienner and Krebs (1991) reported that body weights of red squirrels in spruce and Douglas firhabitats were not influenced by food supplementation despite the presence of excessive food.Consequently, increasing the amount of supplemental food in this study would likely have littleinfluence on current body weight for red squirrels.Alternatively, red squirrels require territories which enclose an adequate food supply tosurvive (C. Smith 1968, Kemp and Keith 1970, Rusch and Reeder 1978). Individuals that loseor do not acquire a territory prior to winter likely will not survive (C. Smith 1968, Kemp andKeith 1970, Rusch and Reeder 1978). M. Smith (1968) reported that red squirrels cachedmany more cones than were necessary to sustain them until the next cone crop. The surpluscones were critical to squirrels during years of poor cone production. Consequently, conecutting and caching is critical for current and future over-winter survival. Middens inlodgepole pine stands are smaller compared to middens in other conifer forests (Hatt 1943,31Gurnell 1984). Gurnell (1984) suggested that the low number of cones cached in middens inlodgepole pine-dominated stands in subarctic and subalpine areas may be the result ofinsufficient time for caching. Therefore, a trade-off may exist between the amount of timespent foraging (to attain a critical or optimum weight) and the time spent caching food. Foodsupplementation may influence red squirrels by reducing the time spent foraging (they canforage more efficiently), thereby increasing the time available to cache cones. The influenceof supplemental feeding, therefore, would be reflected in improved survival rates and bodyweights during years of poor cone production.Since cones remain on lodgepole pine trees (C. Smith 1968), one may argue thatmidden size would not be critical to the survival of red squirrels in these stands. However,when feeding in the open during winter, red squirrels expend larger amounts of energy toobtain food, have increased cost of thermoregulation, and increased exposure to predators,than remaining stationary and feeding in middens (Rusch and Reeder 1978). Consequently,red squirrels with poor cone caches in lodgepole pine stands may have high over-wintermortality. Therefore, cone cutting and caching may be critical to red squirrels in lodgepolepine stands, despite the persistence of lodgepole pine cones on the trees.These results do not support the hypothesis. Population parameters (reproduction,body weight, recruitment, and survival rates) of northern flying squirrels and red squirrelswere not higher in food supplemented stands. A higher rate of food supplementation, coupledwith monitoring red squirrel populations during years of poor cone production, may moreaccurately test this hypothesis. Alternatively, large-scale food supplementation during wintermay have a greater influence than summer food supplementation. Flying squirrels consume alow quality food source (lichen) during the winter (McKeever 1960, Hall 1991). In addition,male red squirrels were significantly heavier on one food supplemented grid than the controlgrid during winter food supplementation (Sullivan 1990). Consequently, reproduction, bodyweight, and survival for flying squirrels and red squirrels may be influenced more duringwinter food supplementation than summer food supplementation.32H3: The abundance of flying squirrels will be positively related to the abundance of cavities.Volts (1986) suggested that the availability of cavities was the primary limitingresource for northern flying squirrels. Observations of flying squirrels’ behaviour does notsupport this prediction. Observation 1: northern flying squirrels do not appear territorial.Flying squirrels both share nests (Cowan 1936, Weigi and Osgood 1974, Maser et al. 1981,Mowrey and Zasada 1982) and interact regularly (Coventry 1932, Cowan 1936). Observation2: the home range of flying squirrels encloses numerous den sites that were used regularly onconsecutive days (Weigi and Osgood 1974, Mowrey and Zasada 1982, Carey 1991, Witt 1992)and occasionally by different flying squirrels. Observation 3: Harestad (1990) erected 80 nestboxes for squirrels. After 4.5 years, only 46 nest boxes were used, of which 22 nests wereidentified as being used by flying squirrels. These observations indicate that there areunoccupied den sites with in a flying squirrels home range. Since flying squirrels do notappear territorial these den sites would be available for occupancy by other flying squirrels.Consequently, den sites likely do not function as a limiting resource for northern flyingsquirrels.The hypothesis that the abundance of flying squirrels will be positively related to theabundance of cavities was not supported by my results. During 45 cavity checks, only 1cavity was occupied by a flying squirrel. All other cavities were unoccupied at the time of thesurvey. A total of 121 squirrels were captured in these stands over a two-year period. Theseobservations indicate that northern flying squirrels do not require, nor strongly prefer, cavitiesas nest sites.33CHAPTER 3POPULATION DYNAMICS OF NORTHERN FLYING SQUIRRELS AN]) REDSQUIRRELS IN SECOND- AND OLD-GROWTH LODGEPOLE PINE STANDSINTRODUCTIONStudies investigating the relative importance of old-growth forests, compared tosecond-growth forests, for northern flying squirrels have yielded inconsistent results. In somestudies flying squirrels were more abundant in old-growth than second-growth forests (Volts1986, Carey 1989, 1991, 1993, Carey et al. 1992, Witt 1992). In contrast, other studiesfound no difference in flying squirrel abundance between old-growth and second-growthforests (Anthony et al. 1987, Aubry et al. 1991, Corn and Bury 1991, Gilibert and Aliwine1991, Rosenberg and Anthony 1992). Since most studies examined flying squirrelpopulations during the fall, comparison of population parameters (body weights, recruitment,survival rates, and proportion of the population breeding) between second-growth and old-growth forests could not be investigated.Red squirrels are a common resident of coniferous forests throughout North America(C. Smith 1981). Mature coniferous stands are considered optimal habitat for red squirrels(Brink and Dean 1966, C. Smith 1968, M. Smith 1968, Rusch and Reeder 1975). Sullivanand Moses (1986) found that unthinned juvenile lodgepole pine stands exhibited severalcharacteristics of a sub-optimal habitat: low proportion of females breeding, low survivalrates, and high fall recruitment. They concluded that juvenile pine stands may provide adispersal sink for juvenile and yearling red squirrels. There was no consistent difference inred squirrel densities between juvenile and mature lodgepole pine stands (Sullivan and Moses1986, Sullivan 1987).34The objective of this study was to compare the characteristics of flying squirrel and redsquirrel populations in second-growth and old-growth lodgepole pine stands. This researchspecifically tests the following hypotheses, phrased as predictions:Hi: Population size, body weight, recuitment, survival rate, and proportion of northern flyingsquirrels in breeding condition will be lower in second-growth than old-growth lodgepolepine stands.H2: Population size, body weight, recruitment, survival rate, and proportion of red squirrelsin breeding condition will be lower in second-growth than old-growth lodgepole pinestands.METhODSDetails of the study area, experimental design, squirrel population dynamics andstatistical analyses were as reported in Chapter 2, with some modifications.Site elevation ranged from 1356 m to 1590 m in the Montane Spruce (MSdm2)andEngelmann Spruce Subalpine Fir (ESSFdC2)(transition area) biogeoclimatic zones (Meidingerand Pojar 1991). Characteristics of the two second-growth stands were those reported forcontrol stands in Chapter 2. Two old-growth (> 120 years old) lodgepole pine stands wereselected for the study. Dominant coniferous species in these stands were lodgepole pine andvarying amounts of subalpine fir, Engelmann spruce, and hybrid spruce. Old-growth standswere selected according to dominance of lodgepole pine greater than 120-years-old,accessibility, and size. Squirrel populations were live-trapped every two weeks from June toOctober, 1992 (10 trap sessions) and May 13 to August 25, 1993 (8 trap sessions).Wilcoxon’s ranked sign test (Sokal and Roif 1981, pp. 448-449) was based on an average35population size for second-growth stands. Due to dissimilarity of old-growth stands,population sizes in these stands were not averaged.RESULTSTrappabilityJolly trappability for flying squirrels and red squirrels were estimated for each standand year separately. Average flying squirrel trappability in second-growth stands (50.1% ±6.3) was significantly less than old-growth stands (67.7% ± 11.0), based on nonoverlappingconfidence intervals. However, this difference was significant (Table 3.1) for only one of theeight pair-wise comparisons. In 1992, female trappability on SG2 (46.2% ± 28.3) wassignificantly less than 0G2 (88.2% ± 11.6).Average trappability for red squirrels was significantly greater in old-growth (77.3% ±6.1) than second-growth stands (66.3% ± 3.9), based on overlapping 95% confidenceintervals. However, none of the eight pair-wise comparisons (Table 3.1) were significantlydifferent.36Table 3.1. Estimates of Jolly trappability for northern flying squirrels and red squirrels insecond-growth and old-growth stands (9 samples/mean; old-growth - 1993, 7samples/mean). Values are mean percent trappability (average (number caught/Jolly-Seberpopulation estimate) x 100%, Krebs and Boonstra 1984) with 95% confidence intervals inparentheses.1992 1993FEMALE MALE FEMALE MALEFLYING SOUIRRELSecond-growth1 55.2 66.7 47.7 34.9(30.6-79.9) (49.5-83.8) (30.0-65.3) (12.8-57.0)2 46.2a 49.4 47.0 53.5(18.0-74.5) (28.6-70.3) (17.6-76.4) (30.3-76.9)Old-growth1 67.4 77.6 71.4 70.9(46.0-88.9) (55.2-99.9) (46.7-96.1) (44.0-77.7)2 88.2a 71.2 33.6 61.4(76.7-99.8) (59.0-83.5) (12.2-54.9) (38.2-84.7)RED SQUIRRELSecond-growth1 63.0 66.9 58.7 71.7(44.1-81.9) (53.7-80.1) (36.1-81.2) (55.3-88.0)2 65.7 63.7 64.1 76.7(45.7-85.7) (48.1-79.2) (45.0-83.2) (64.1-89.3)Old-growth1 78.8 85.2 88.7 76.2(62.7-94.9) (72.0-98.4) (70.9-106.5) (42.0-110.6)2 85.7 66.0 67.1 70.4(74.7-96.8) (53.1-78.9) (32.1-102.1) (53.5-87.4)a - a comparisons were significantly different, based on the 95% confidence intervals37Population densityThere was no consistent difference between estimated population sizes for flyingsquirrels (Figure 3.1) in second-growth (SG) and old-growth (OG) stands. Wilcoxon’s rankedsign test indicated that the average population size in second-growth stands was significantlyhigher than OG1 (P = 0.002) but significantly less than 0G2 (P = 0.02). These results weresimilar when population sizes were based on resident squirrels only (SG > OG1, P = 0.001;SG < 0G2, P = 0.009). Population sizes of flying squirrels in second-growth stands werenot significantly different from each other (P = 0.51), while population sizes in old-growthstands were (0G2 > OG1, P = 0.001). Average densities of flying squirrels in second-growth stands were; SG1: 0.64 squirrels/ha, SG2: 0.68 squirrels/ha (MNA: 0.57 and 0.54squirrels/ha, respectively). There was a three-fold increase in flying squirrel densities betweenOG1 (0.35 squirrels/ha) and 0G2 (1.03 squirrels/ha) (MNA: 0.31 and 0.85 squirrels/ha,respectively).I further examined the population size of flying squirrels in second-growth and old-growth stands for each year separately. Average population sizes of flying squirrels in second-growth stands were significantly higher than OG1 in 1992 (P = 0.03) and 1993 (P = 0.03).These results are the same when population estimates were based on resident squirrels (1992:SG > OG1, P = 0.02; 1993: SG > OG1, P 0.02). In 1992, average population sizes inSG stands were lower than 0G2 (P = 0.02); in 1993, they were not significantly different (P= 0.09). Population estimates, based on resident squirrels, in 1992 for second-growth standswere lower than 0G2 (P = 0.02); in 1993, they were not significantly different (P = 0.31).Differences in red squirrel densities between second-growth and old-growth stands wereinconsistent (Figure 3.2). Wilcoxon’s ranked sign test indicated that population sizes insecond-growth stands were significantly higher than old-growth stands (SG > OG1, P =0.004; SG > 0G2, P = 0.03). Population estimates, based on resident squirrels, in secondgrowth stands were significantly higher than OG1 (P = 0.004), but not significantly different38from 0G2 (P = 0.28). Red squirrel population sizes in second-growth stands were notsignificantly different from each other (P = 0.20). Population sizes in old-growth stands weresignificantly different from each other (0G2 > OG1, P < 0.01). The estimated populationsize in second-growth replicates, based on resident squirrels, were significantly different (SG2> SG1, P = 0.01), as were the old-growth replicates (P < 0.01).Average densities for red squirrels were: SG1: 1.58 squirrels/ha, SG2: 1.71squirrels/ha, OG1: 1.21 squirrels/ha, 0G2: 1.59 squirrels/ha. Average densities for redsquirrels, based on MNA, were; SG1: 1.35 squirrels/ha, SG2: 1.55 squirrels/ha, OG1: 1.11squirrels/ha, 0G2: 1.43 squirrels/ha.RecruitmentThe number of recruits for flying squirrels in second-growth stands (average for 16 trapperiods ± 95% CI; 0.20 ± 0.26 and 0.50 ± 0.79) did not differ from old-growth stands(0.63 ± 0.52 and 0.63 ± 0.52) (P = 0.68). The number of recruits for red squirrels insecond-growth stands (1.00 ± 1.49 and 1.70 ± 0.97) did not differ from old-growth stands(0.88 ± 0.58 and 1.63 ± 0.90) (P = 0.08).39Figure 3.1. Estimated population sizes of flying squirrels for second- and old-growth stands. Population estimates based on all individuals captured andresident individuals captured are shown.A. ALL INDIVIDUALS.I)0r1NC,,zCC1992SECOND-GROWTH 1- SECOND-GROWTH 2B. RESIDENTS1993-*- OLD-GROWTH 1OLD-GROWTH 2J J A S AM J J A1993— SECOND-GROWTH 1 * OLD-GROWTH 1SECOND-GROWTH 2 0 OLD-GROWTH 240010IIA. ALL INDIVIDUALS1992SECOND-GROWTH 1-4- SECOND-GROWI’H 21993B. RESIDENTSt OLD-GROWJII 1-9- OLD-GROWTH 2— SECOND-GROWTH 1.44. SECOND-GROWTH 2J S AS AM J J A1992 1993-*- OLD-GROWTH 1-9-- OLD-GROWTH 2Figure 3.2. Estimated population sizes of red squirrels for second- and old-growthstands. Population estimates based on all individuals captured and residentindividuals captured are shown.41Body weightAlthough adult male flying squirrels were heavier (Figure 3.3 A) in second-growth(mean ± 95% CI; SG1: 139.0 g ± 4.2, SG2: 139.7 g ± 4.2) than old-growth stands (OG1:133.0 g ± 16.1, 0G2: 131.2 g ± 6.5) in 1992, the difference was not significant (ANOVA,P 0.31). In 1993, average weights of adult male flying squirrel were not significantlydifferent in second-growth stands (SG1: 146.4 g ± 10.4, SG2: 142.6 g ± 8.1) and old-growthstands (OG1: 143.3 g ± 28.3, 0G2: 140.2 g ± 12.0) (ANOVA, P = 0.79).In 1992, the average weight of adult male red squirrels (Figure 3.3B) were notsignificantly different in second-growth (SG1: 237.4 g ± 11.2, SG2: 229.3 g ± 12.1) andold-growth stands (OG1: 226.4 g ± 6.4, 0G2: 251.9 g ± 14.0) (ANOVA, P = 0.72).Although average weights of adult male red squirrels, in 1993, were heavier in old-growthstands (OG1: 250.0 g ± 14, 0G2: 245.0 g ± 6.3) than second-growth stands (SG1: 229.0 g± 9.4, SG2: 235.3 g ± 8.8), the difference was not significant (ANOVA, P = 0.95).Survival ratesSurvival rates, based on 28-day intervals, were calculated for summer 1992, late fall,over-winter 1992 - 1993, and spring 1993, to determine if survival rates varied betweensecond-growth and old-growth stands. Survival rates for flying squirrels (Figure 3.4A) andred squirrels (Figure 3.4B) were not significantly different during each period (based onnonoverlapping 95% confidence intervals) for second- and old-growth stands.42A) NORTHERN FLYING SQUIRRELSFigure 3.3. Body weight of adult male flying squirrels (A) and red squirrels (B).Values represent average weight with 95% confidence intervals for second-growth(SG1 and SG2) and old-growth (OG1 and 0G2) stands. Sample sizes are recordedabove the upper confidence interval.1803170160130120110100I I I I I I I I I I ISi S2 01 02 Si S2 01 021992 1993B) RED SQUIRRELS270F$17260250U19240230220210 I I I I I I I I I I I I I1992Si S2 01 02 Si S2 01 02199343A) NORTHERN FLYING SQUIRRELSFigure 3.4. Flying squirrel (A) and red squirrel (B) survival rates for second-growth(SG) and old-growth (OG) stands. Survival rates, based on 28-day intervals, werecalculated for four periods: summer 1992, fall, over-winter, and spring 1993.Values are 28-day survival rates with 95% confidence intervals.21.510.5 iDHJiC,,Si S2 0102 Si S2 0102 Si S2 0102 Si S2 0102B) RED SQUIRRELSSi $20102 Si S2 0102 Si S2 0102 Si S2 0102SUMMER FALL WINTER SPRING1992 1992-93 199344BreedingThe percentages of adult flying squirrels in breeding condition in second- and old-growth stands are shown in Table 3.2. Overall, significantly more female flying squirrels bredin old-growth than second-growth stands (x2 = 7.98, df =3, 0.025 < P < 0.05). Theproportion of female flying squirrels breeding in 1992 was not significantly different insecond-growth and old-growth stands (x2 = 0.64, df =1, 0.25 < P < 0.50). However, in1993, significantly more female flying squirrels bred in old-growth than second-growth stands(x2 = 7.34, df = 1, 0.005 < P < 0.01). The average proportion of females breeding(across all stands and years) was 40.9%. A significantly greater proportion of male flyingsquirrels bred in second-growth (SG1: 63.6%, SG2: 63.0%) than old-growth stands (OG1:50.0%, OG2: 40.0%) (x2 = 6.70, df = 1, 0.005 < P < 0.01).Fewer female red squirrels bred in second-growth (average for 1992: 10.7%, 1993:37.1%) than old-growth stands (average for 1992: 25.7%, 1993: 70.8%) (x2 = 64.3, df = 3,P < 0.005) (Table 3.2). The proportion of female red squirrels breeding was significantlygreater in old-growth than second-growth stands in 1992 (x2 = 32.7, df = 1, P < 0.005) andin 1993 (x2 = 31.6, df = 1, P < 0.005). The proportion of male red squirrels breeding insecond-growth stands was not significantly different from old-growth stands (x2 = 2.16, df =1, 0.10 < P < 0.25).45Table 3.2. Percentage of adult northern flying squirrels and red squirrels in breedingcondition in second-growth and old-growth stands for 1992 and 1993 (sample size inparentheses).1992 1993FEMALE MALE FEMALE MALEFLYING SOUIRRELSecond-growth11 26.3 0 391a 63.6(19) (30) (23) (11)2 37.5 0 438a 63.0(16) (22) (16) (27)Old-growth1 27.8 0 70•0b 50.0(18) (14) (10) (12)2 36.8 0 632b(38) (43) (19) (25)RED SQUIRREL2Second-growth1 0 468C 82.8(35) (58) (47) (64)2 128c 0 91.3(86) (52) (50) (80)Old-growth1 357d 0 714d 88.0(42) (95) (35) (25)2 194d 0 700d 80.8(67) (83) (30) (52)1) proportion of female flying squirrels breeding in 1993 was significantly higher in old-growth stands than second-growth stands, (x2 = 7.34, df = 1, 0.005 < P < 0.01)2) proportion of female red squirrels breeding in second-growth stands was significantly lower than old-growth stands, (x264.3, df = 3, P < 0.005)46MovementMean maximum distance moved (MMDM) between consecutive trap periods are shownin Table 3.3. The smallest and largest MMDM traveled both for flying squirrels and redsquirrels occurred in second-growth stands. Average MMDM for flying squirrels in second-growth and old-growth stands were 93.6 m and 92.7 m, respectively.The MMDM in second-growth stands for red squirrels was twice as great in 1993 than1992. Average MMDM in second-growth stands in 1992 and 1993 was 59.0 m and 123.6 m,respectively. MMDM was more consistent in old-growth stands. In 1992, the averageMMDM in old-growth was 75.2 m, while in 1993 the average MMDM was 97.9 m.47Table 3.3. Movement and mean maximum distance moved (MMDM) between consecutive capturesfor flying squirrels and red squirrels (MMDM = mean distance moved + 3 standard deviations).number of mean distance standard MMDMobservations moved (m) deviation (m)FLYING SOUIRRELSSecond-growth 11992 2 111.1 4.78 125.51993 8 44.6 2.49 52.1Second-growth 21993 6 77.8 8.54 103.3Old-growth 11993 5 94.1 6.69 114.2Old-growth 21992 23 69.5 7.5 92.11993 6 54.1 5.86 71.7RED SOUIRRELSSecond-growth 11992 27 40.1 6.03 58.21993 28 104.2 9.14 131.6Second-growth 21992 61 42.3 5.8 59.71993 30 100.4 8.4 125.6Old-growth 11992 60 51.1 7.6 74.01993 26 87.5 8.2 112.1Old-growth 21992 77 54.6 7.3 76.41993 11 63.5 6.7 83.648DISCUSSIONSecond-growth stands have been classified as sub-optimal habitat for northern flyingsquirrels (Volts 1986, Carey 1989, Witt 1992) and red squirrels (Sullivan and Moses 1986).The objective of this study was to compare the characteristics of flying squirrel and redsquirrel populations in second-growth and old-growth lodgepole pine stands.Northern flying squirrels and red squirrels were trapped for two summers in twosecond-growth and two old-growth lodgepole pine stands. In general, populationcharacteristics (density, weight, recruitment, survival, and MMDM) of flying squirrels and redsquirrels were similar between second-growth replicates. Therefore, I am confident that thepopulation parameters measured in second-growth stands, during this study, accurately reflectthe true population parameters for this stand type. It should be noted that the Wilcoxon’sranked sign tests were based on population estimates repeated over time for the same stand,and may be construed as temporally pseudoreplicated (Huribert 1984).The availability of old-growth stands was limited. The two old-growth sites used forthis study were the best sites available based on accessibility, size, and dominance of locigepolepine greater than 120-years-old. Both sites were contained within the same old-growth standand separated by approximately 400 meters. One old-growth replicate (OG2) was situatedwithin a continuous stand of old-growth lodgepole pine. To maximize the distance betweengrids, the first site (OG1) was placed within a narrow segment of old-growth protruding offthe continuous stand. As a result, OG1 was confined by recent clear cuts. Both longitudinalborders of the 6 X 16 grid for OG1 were within 100 meters of the stand edge. The lowerdensity of flying squirrels in OG1 may have resulted from a smaller effective trapping area,coupled with a lower potential for immigration. Consequently, population size and densityestimates for flying squirrels in OG2 may more accurately represent true values for old-growthlodgepole pine stands. It is important to increase site replication, in future studies, to reduceinconsistencies among replicates.49Sullivan and Moses (1986) reported similar densities of red squirrels in second-growthand mature stands. However, low proportion of females in breeding condition, low survival,and high recruitment in second-growth stands indicated that these stands were sub-optimalhabitat for red squirrels. Consequently, habitat quality cannot be gauged by density valuesalone (Van Home 1983). To determine if second-growth lodgepole pine stands were sub-optimal habitat for northern flying squirrels and red squirrels, two hypotheses were tested.Hi: Population size, body weight, recruitment, survival rate, and proportion of northernflying squirrels in breeding condition will be lower in second-growth than old-growthlodgepole pine stands.Previously reported densities of flying squirrels in second-growth and old-growthstands are highly variable. Rosenberg and Anthony (1991) reported the highest densities: 2.0squirrels/ha in second-growth, and 2.3 squirrels/ha in old-growth stands. Flying squirreldensities in northern Washington averaged 0.2 squirrels/ha in second-growth and 0.5squirrels/ha in old-growth stands (Carey et al. 1992). Intermediate flying squirrel densitieswere reported for western Oregon: second-growth - 0.12 squirrels/ha, old-growth - 0.85squirrels/ha (Witt 1992); second-growth - 0.9 squirrels/ha, old-growth stands - 2 squirrels/ha(Carey et a!. 1992).My results indicated that old-growth lodgepole pine stands do not consistently supporthigher densities (0.35 and 1.03 squirrels/ha) of flying squirrels than second-growth stands(0.64 and 0.68 squirrels/ha). Second-growth 1 had significantly more flying squirrels thanold-growth 1. However old-growth 2 was considered more representative of an old-growthstand, since it was placed within a continuous stand of old-growth lodgepole pine. Old-growth2 had a significantly higher population size of flying squirrels than second-growth 2, in 1992.In 1993, population sizes of flying squirrels in second-growth stands were not significantly50different from old-growth 2. Flying squirrel abundance within stands appeared to fluctuatewidely.If second-growth stands were sub-optimal habitat for flying squirrels, lower bodyweights, recruitment, survival rates, proportion of flying squirrels in breeding condition, and agreater average MMDM would be expected. In this study, average adult male weight,recruitment, survival rates, proportion of females breeding in 1992, and MMDM were notsignificantly different in second-growth and old-growth stands. The proportion of femaleflying squirrels in breeding condition in 1993 was significantly lower in second-growth thanold-growth lodgepole pine stands. The proportion of male flying squirrels in breedingcondition, in 1993, was significantly greater in second-growth stands than old-growth stands.Overall, these results do not support the hypothesis. Consequently, these results wouldindicate that, for northern flying squirrels, second-growth lodgepole pine stands were notlower in quality than old-growth lodgepole pine stands during this study.H2: Population size, body weight, recruitment, survival rate, and proportion of red squirrelsin breeding condition will be lower in second-growth than old-growth lodgepole pinestands.My study indicated that population sizes of red squirrels were larger in managedsecond-growth than old-growth lodgepole pine stands. The average densities in second-growthand old-growth stands were 1.65 and 1.40 squirrels/ha, respectively. These densities arecomparable to other studies reporting densities of red squirrels in pine dominated forests: 1.1squirrels/ha (extrapolated from territorial size in C. Smith 1968), 0.86 - 2.6 squirrels/ha(Rusch and Reeder 1978), 1.3 squirrels/ha (Gurnell 1984), 1.1 - 1.3 squirrels/ha (Sullivan andMoses 1986). Sullivan and Moses (1986) found similar densities (1.1 squirrels/ha) of redsquirrels in unthinned and mature lodgepole pine stands. However, they reported lowerdensities of red squirrels in thinned second-growth (0.22 and 0.54 squirrels/ha) than mature51old-growth lodgepole pine stands. In addition, red squirrels in unthinned stands had lowersurvival rates, lower female reproduction, and higher recruitment than mature stands. Incontrast, I found that population sizes were higher in second-growth than old-growth locigepolepine stands. In addition, adult male body weights, recruitment, survival, MMDM, andproportion of males in breeding condition for red squirrels did not differ in thinned second-growth and old-growth lodgepole pine stands. The proportion of female red squirrels breedingwas lower in second-growth than old-growth stands.Kemp and Keith (1970) noted that an increase in female reproduction occurred before agood cone crop. They suggested that female reproduction is less dependent upon current foodavailability and more dependent upon anticipated cone production. The spruce and fir conecrop for 1993 was the highest recorded cone crop in the region since 1989 (Don Purdy,Ministry of Forests, Vernon, B. C., personal communication). Engelmann spruce andsubalpine-fir are the climax species on these sites, and hence old-growth stands had a greaterproportion of spruce and fir than second-growth stands (personal observation). If femalereproduction is tied to anticipated cone production, then a greater proportion of female redsquirrels breeding in old-growth stands than second-growth stands, in 1993, would beexpected.If second-growth stands were sub-optimal habitat for red squirrels, lower densities,body weights, recruitment, survival rates, proportion of red squirrels in breeding condition,and a greater average MMDM would be expected. However, I found that densities, adultmale body weights, recruitment, survival, MMDM, and proportion of males in breedingcondition for red squirrels were either higher or not significantly different in thinned secondgrowth and old-growth lodgepole pine stands. These results do not support the hypothesis thatsecond-growth lodgepole pine stands were sub-optimal habitat for red squirrels.52CHAPTER 4SUMMARYThe two main objectives of this thesis were: 1) to examine population dynamics andfood limitation of northern flying squirrels and red squirrels (Chapter 2) and 2) to determine ifsecond-growth forests were sub-optimal habitat for flying squirrels and red squirrels (Chapter3). In chapter 2, I tested the hypotheses that flying squirrels and red squirrels were limited byfood availability, and that flying squirrel abundance was positively related to the abundance ofcavities.The average weight of adult males, recruitment, survival rates, and the proportions offemale and male flying squirrels, and male red squirrels in breeding condition were notsignificantly different in control and treatment stands. The proportion of female red squirrelsin breeding condition was significantly higher in control than treatment stands. Consequently,population size of flying squirrels and red squirrels appeared to be limited by food availability,but individual squirrels did not appear to benefit from food supplementation.These results are consistent with similar studies examining food limitation in arborealsciurids (Sullivan and Sullivan 1982, Sullivan 1990, Klenner and Krebs 1991, Sullivan andKlenner 1993). These studies suggested that the poor response of individual squirrels to foodsupplementation may have resulted from limitations in experimental design. In the presentstudy I adjusted the experimental design to compensate for these limitations. The resultsremained the same: individual squirrels did not appear to benefit from food supplementation.Sunflower seeds, in seed plots, were completely consumed or removed to cache sitesprior to the next application (Kohier 1993). Therefore, food may still have been limitedduring this study. The lack of a response to supplemental food by individual squirrels mayhave reflected this fact. However, flying squirrels consume a lower quality food source duringwinter (McKeever 1960, Hall 1991), while red squirrels have an increased cost of53thermoregulation (Rusch and Reeder 1978). Therefore supplemental feeding during wintermonths may have a greater influence on individual squirrels.Alternatively, the proportion of female red squirrels in breeding condition may not beinfluenced by current food availability. Kemp and Keith (1970) suggested that femalereproduction was less dependent upon current food availability and more dependent uponanticipated cone production. This hypothesis may explain the inability of supplemental feedingto increase female red squirrel reproduction. Female flying squirrel reproduction may also bedependent upon anticipated natural food production, however, not enough information existsconcerning flying squirrel biology to support or refute this hypothesis.Gurnell (1984) suggested that red squirrels may have insufficient time in subarctic andalpine areas to cache cones. Therefore, a trade-off may exist between the amount of timespent foraging and the time spent caching food. Similarly, maneuverability and glidingefficiency for flying squirrels is influenced by body weight (Thorington and Heaney 1981).Wauters and Dhondt (1989) found that survival was positively correlated with body weight inred squirrels (Sciurus vulgaris) in Belgium. Therefore, the poor influence of supplementalfeeding on flying squirrel and red squirrel body weights (and concomitant survival rates) mayreflect the optimization of body weight with gliding efficiency and cone caching, respectively.Future studies should extend the food supplementation period to include winter months,coupled with behavioural observations. For example, flying squirrels typically exhibit abiphasic activity pattern (Weigl and Osgood 1974, Mowrey and Zasada 1882). Animals lefttheir nest shortly after dark and returned after 2 hours (average 118 mm.). Their secondactivity period, near dawn, was shorter (76 mill.) (Weigl and Osgood 1974). If flyingsquirrels were optimizing their weight, these activity periods would be shorter in foodsupplemented stands than control stands. If red squirrels in subarctic and subalpine areas wereoptimizing their weight, foraging periods would be shorter and cone caches larger in foodsupplemented stands than control stands.54Food abundance and cavity availability have been suggested as a potential limitingresources for flying squirrels (Carey 1991, Carey et a!. 1992, Volts 1986). The availability ofcavities was considered the primary limiting resource for northern flying squirrels (Volts 1986,Carey 1991, Carey, et a!. 1992). This hypothesis was not supported by my results. During45 cavity checks, only one cavity was occupied by a flying squirrel. Flying squirrels are themain food source for spotted owls (Strix occidentalis) (Forsman et a!. 1984, Carey et a!.1992). Consequently, it has been suggested that providing cavities in managed stands mayraise flying squirrel populations, thus, increasing available foraging habitat for spotted owls(Carey unpublished). My results indicated that improving food availability may have a greaterinfluence on population sizes of flying squirrels than improving cavity availability. However,this conclusion is based on a limited number of observations.Future studies should more intensively examine the importance of cavities in limitingflying squirrel populations. In particular, these studies should be conducted in habitatsoccupied by spotted owls.In chapter 3, I tested the hypothesis that second-growth stands were sub-optimalhabitat, relative to old-growth stands, for flying squirrels and red squirrels. If true, lowerbody weights, recruitment, survival rates, proportion of flying squirrels in breeding condition,and a greater average MMDM would be expected. However, population sizes of flyingsquirrels, body weight, recruitment, and survival of flying squirrels and red squirrels were notsignificantly different in second-growth and old-growth stands. Population sizes of redsquirrels were larger in second-growth than old-growth stands. There was no consistentdifference in the proportion of squirrels in breeding condition between second-growth and old-growth stands. These results indicated that second-growth stands were not lower in qualitythan old-growth stands for flying squirrels and red squirrels during this study.This is the first study, that I am aware of, that critically examines habitat preferences ofnorthern flying squirrels using population dynamics (body weight, recruitment, reproduction,and survival rates) in addition to density. It is one of three studies critically examining habitat55preference of red squirrels using population dynamics. Using population dynamics to gaugehabitat quality is more reliable than using density alone. In this study, both populationdynamics and density are consistent in failing to support the hypothesis that second-growthstands were suboptimal habitat for flying squirrels and red squirrels. However, The scope ofthis study was limited both spatially and temporally. Thus, due to the limited number ofsimilar studies, generalization are limited both spatially and temporally.Management ImplicationsMethods of increasing population sizes of squirrels may be important in managing for‘Threatened’ or ‘Endangered’ wildlife species directly (endangered Mount Graham red squirrel(T. hudsonicus grahamensis. ) or indirectly (northern flying squirrels -- spotted owls).Supplemental feeding has been considered for managing the endangered Mount Graham redsquirrel population during periods of low population size and presumed nutritional stress(Koford 1992). Enhancing northern flying squirrel populations may be important in increasingforaging habitat for spotted owls (Carey unpublished). Northern flying squirrel and redsquirrel populations appear to be limited by the availability of food. Consequently, wildlifemanagement programs that increase food availability for northern flying squirrels and redsquirrels will likely increase their population size and successfully meet the managementprogram objectives.The lack of a consistent pattern in flying squirrel and red squirrel population parametersbetween second-growth and old-growth stands indicated that this relationship was not a simpleone. Further studies must be conducted to identify the relationships (resource limitations orregulating mechanisms) between northern flying squirrel and red squirrel populations withvarious forest characteristics and ages. Old-growth forests act as scientific controls in whichthese relationship exist undisturbed. Therefore, it is critical to maintain large undisturbedstands of old-growth forests to improve our understanding of these relationships.56LITERATURE CITEDANTHONY, R. G., E. D. FORSMAN, G. A. GREEN, G. WITMER, and S. K.NELSON. 1987. Small mammal populations in riparian zones of different-agedconiferous forests. The Murrelet, 68:94-102.AUBRY, K. B., M. CRITES, and D. WEST. 1991. Regional patterns of small mammalabundance and community composition in Oregon and Washington. Pages 284-294 inL. F. Ruggiero, K. B. Aubry, A. B. Carey, and M. Huff, technical coordinators,Wildlife and vegetation ofunmanaged Douglas-firforests. U.S.D.A. Forest Service,General Technical Report, PNW-285. 11 pp.BOUT1N, S. 1990. Food supplementation experiments with terrestrial vertebrates: patterns,problems, and the future. Canadian Journal of Zoology, 68: 203-220.BRINK, C. H., and F. C. DEAN. 1966. Spruce seeds as a food of red squirrels and flyingsquirrels in interior Alaska. Journal of Wildlife Management, 30:503-5 12.CAREY, A. B. (Unplished). Experimental manipulation of managed stands to provide habitatfor spotted owls and to enhance plant and animal diversity: a summary andbackground for the USDA FS experiment on Ft. Lewis, Washington. PacificNorthwest Research Station, USDA Forest Service, Olympia, Washington. 8 pp.CAREY, A. B. 1989. Wildlife associated with old-growth forests in the Pacific Northwest.Natural Areas Journal, 9(3):151-162. 46 pp.CAREY, A. B. 1991. The biology of arboreal rodents in Douglas-fir forests. U.S.D.A.Forest Service, Pacific Northwest Research Station, General Technical Report PNWGTR-276.CAREY, A. B., B. L. BISWELL, and J. W. WITT. 1991a. Methods for measuringpopulations of arboreal rodents. U.S.D.A. For. Serv. Gen. Tech. Rep. PNW GRT273. 24 pp.CAREY, A. B., M. M. HARDT, S. P. HORTON, and B. L. BISWELL. 1991b. Springbird communities in the Oregon Coast Ranges. Pages 123-144 in L. F. Ruggiero, K.B. Aubry, A. B. Carey, and M. Huff, technical coordinators, Wildlife and vegetationofunmanaged Douglas-firforests. U.S.D.A. Forest Service, General TechnicalReport, PNW-285.CAREY, A. B., S. P. HORTON, and B. L. BISWELL. 1992. Northern spotted owls:influence of prey base and landscape character. Ecological Monographs, 62:223-250.57CORN, P. S., and R. B. BURY. 1991. Small mammal communities in the Oregon CoastRange. Pages 240-254 in L. F. Ruggiero, K. B. Aubry, A. B. Carey, and M. Huff,technical coordinators, Wildlife and vegetation ofunmanaged Douglas-firforests.U.S.D.A. Forest Service, General Technical Report, PNW-285.COVENTRY, A. F. 1932. Notes on the Mearns flying squirrel. Canadian Field-Naturalist,46:75-78.COWAN, I. McT. 1936. Nesting habits of the flying squirrel Glaucomys sabrinus. Journalof Mammalogy, 17:58-60.DEMPSEY J. A., and D. M. KEPPIE. 1993. Foraging patterns of eastern red squirrels.Journal of Mammalogy, 74:1007-1013.FORSMAN, E. D., MESLOW, E. C., and WIGIIT, H. M. 1984. Distribution and biologyof the spotted owl in Oregon. Wildlife Monographs No. 87.GILBERT, F. F., and R. ALLWINE. 1991. Small mammal communities in the OregonCascade Range. Pages 256-267 in L. F. Ruggiero, K. B. Aubry, A. B. Carey, andM. Huff, technical coordinators, Wildlife and vegetation of unmanaged Douglas-firforests. U.S.D.A. Forest Service, General Technical Report, PNW-285.GURNELL, J. 1983. Squirrel numbers and the abundance of tree seeds. Mammal Review,13:133-148.GTJRNELL, J. 1984. Home range, territoriality, caching behaviour and food supply of thered squirrel (Tamiasciurus hudsonicusfremonti) in a subalpine lodgepole pine forest.Animal Behaviour, 32:1119-1131.HALL, D. S. 1991. Diet of the northern flying squirrel at Sagehen Creek, California. Journalof Mammalogy, 72:615-617.HARESTAD, A. S. 1990. Nest site selection by northern flying squirrels and Douglassquirrels. Northwest Naturalist, 71:43-45.HATT, R. T. 1943. The pine squirrel in Colorado. Journal of Mammalogy, 24:311-345.IIURLBERT, S. H. 1984. Pseudoreplication and the design of ecological field experiments.Ecological Monographs 54:187-211.JIURLY, T. A., and R. J. ROBERTSON. 1987. Scatterhoarding by territorial red squirrels:a test of the optimal density model. Canadian Journal of Zoology, 65:1247-1252.JOLLY, G. M. 1965. Explicit estimates from capture-recapture data with both death andimmigration - stochastic model. Biometrika, 52:225-247.58JOLLY, G. M., and J. M. DICKSON. 1983. The problem of unequal catchabiity in mark-recapture estimation of small mammal populations. Canadian Journal of Zoology,61:922-927.KEMP, G. A. and L. B. KEITh. 1970. Dynamics and regulation of red squirrel(Tamiasciurus hudsonicus) populations. Ecology, 5 1:763-779.KLENNER, W. 1991. Red squirrel population dynamics. II. settlement patterns andresponse to removals. Journal of Animal Ecology, 60:979-993.KLENNER, W., and C. J. KREBS. 1991. Red squirrel population dynamics. I. The effectof supplemental food on demography. Journal of Animal Ecology, 60:961-978.KOFORD, R. R. 1992. Does supplemental feeding of red squirrels change population size,movement, or both? Journal of Mammalogy, 73(4):930-932.KOULER, C. 1993. Influence of large-scale food supplementation on diversity of rodentcommunities. M.Sc. Thesis, University of British Columbia, Vancouver, B.C. 62 pp.KOPROWSKI, J. L. 1991. Response of fox squirrels and grey squirrels to a late spring -early summer food shortage. Journal of Mammalogy, 72:367-372.KREBS, C. J. 1966. Demographic changes in fluctuating populations of Microtusca4fornicus. Ecological Monographs, 36:239-273.KREBS, C. J., and R. BOONSTRA. 1984. Trappability estimates for mark-recapture data.Canadian Journal of Zoology, 62:2440-2444.LACK, D. 1954. The Natural Regulation ofAnimal Numbers. Clarendon Press, Oxford. 343pp.LARSEN, K. W. 1990. Maternal and environmental factors influencing reproductive successof female red squirrels, Tamiasciurus hudsonicus. Progress/candidacy report.Department of Zoology, University of Alberta, Alberta, Canada. 21 pp.MASER, C., R. ANDERSON, and E. L. BULL. 1981. Aggregation and sex segregation innorthern flying squirrels in northeastern Oregon, an observation. Canadian Journalof Zoology, 63:1084-1088.MASER, C., Z. MASER, J. WITT, and G. hUNT. 1978. The northern flying squirrel: amycophagist in southwestern Oregon. Canadian Journal of Zoology, 64:2086-2089.MASER, C., R. 0. ANDERSON, K. CROMACK, JR., (and others). 1979. Dead anddown woody material. In Thomas, J. W., tech. ed. Wildlife habitats in managed59forest: the Blue Mountains of Oregon and Washington, Agric. handb. 553.Washington, D.C.: U. S. Department of Agriculture, pp. 78-95.McKEEVER, S. 1960. Food of the northern flying squirrel in northeastern California.Journal of Mammalogy, 41:270-271.MEIDINGER, B., and J. POJAR. 1991. Ecosystems ofB.C. Ministry of Forests, ResearchBranch, Special Report Series no. 6. Crown Pubi. Inc., Victoria, British Columbia.330 pp.MENDENHALL, W. 1983. Introduction to Probability and Statistics, Sixth edition,Doxbury Press, Boston. 646 pp.MOWREY, R. A., and J. C. ZASADA. 1984. Den tree use and movements of northernflying squirrels in interior Alaska and implications for forest management. In Fishand Wildlife Relationships in Old-growth Forest: proceedings of a symposium (April,1982, Juneau Alaska) (W. R. Meehan, T. R. Merrell, Jr., and T. A. Hanley, eds.),BookMaster, Ashland, Ohio. 425 pp.ROSENBERG, D. K., and R. ANThONY. 1990. Characteristics of northern flying squirrelpopulations in young- and old-growth forests of western Oregon. Canadian Journalof Zoology, 70:161-166.RUSCH, D. A. and W. G. REEDER. 1978. Population ecology of Alberta red squirrels.Ecology, 59:400-420.SEBER, G. A. F. 1982. The Estimation ofAnimal Abundance and Related Parameters,second edition, Charles Griffin and Company, London. 654 pp.SINCLAIR, A. R. E. 1989. Population regulation in animals. In Ecological concepts, thecontribution of ecology to an understanding of the natural world (Ed. by J. M.Cherrett), pp. 197-241. Blackwell Scientific, Oxford.SMITh, C. C., 1968. The adaptive nature of social organization in the genus of treesquirrels Tamiasciurus. Ecological Monographs, 38:31-63.SMITH, C. C., 1981. The indivisible niche of Tamiasciurus: an example of nonpartitioningof resources. Ecological Monographs, 51:343-363.SMITH, M. 1968. Red squirrel responses to spruce cone failure in interior Alaska. Journalof Wildlife Management, 32:305-317.SOKAL, R. R., and F. J. ROHLF. 1981. Biometiy, second edition, W. H. Freeman andCompany, New York. 859 pp.60SULLIVAN, T. P. 1990. Responses of red squirrel (Tamiasciurus hudsonicus) populationsto supplemental food. Journal of Mammalogy, 71:579-590.SULLIVAN, T. P., and W. KLENNER. 1993. Influence of diversionary food on redsquirrel populations and damage to crop trees in young lodgepole pine forest.Ecological Applications, 3:708-7 18.SULLIVAN, T. P., and R. A. MOSES. 1986. Red squirrel populations in natural andmanaged stands of lodgepole pine. Journal of Wildlife Management, 50:595-601.SULLIVAN, T. P., and D. S. SULLIVAN. 1982. Population dynamics and regulation ofthe Douglas squirrel (Tamiasciurus douglasii) with supplemental food. Oecologia,53:264-270.SULLIVAN, T. P. 1987. Red squirrel population dynamics and feeding damage in juvenilestands of lodgepole pine. Research Branch, Ministry of Forest, Victoria, BritishColumbia Forest Research Development Agreement Report, 019: 20 p.TIIORINGTON, R. W. , Jr., and L. R. HEANEY. 1981. Body proportions and glidingadaptations of flying squirrels (Petauristinae). Journal of Mammalogy, 62:101-114.VAN HORN, B. 1983. Density as a misleading indicator of habitat quality. Journal ofWildlife Management, 47:893-901.VOLTS, K. 1986. Habitat requirements of northern flying squirrels in west-central Oregon.M.Sc. Thesis Washington State University, Pullman, WA.WATSON, A. and R. MOSS. 1970. Dominance, spacing behaviour and aggression inrelation to population limitation in vertebrates. In Animal Population in Relation toTheir Food Resources (ed. by A. Watson), pp. 167-220, Blackwell ScientificPublications, Oxford.WATSON, A. (1970). Animal population in relation to theirfood resources. BlackwellScientific Publications, Oxford.WAUTERS, L., and A. A. DHONI)T. 1989. Body weight, longevity and reproductivesuccess in red squirrels (Sciurus vulgaris). Journal of Animal Ecology, 58:637-65 1.WEIGL, P. D. and D. W. OSGOOD. 1974. Study of the northern flying squirrel,Glaucomys sabrinus, by temperature telemetry. American Midland Naturalist,92:482-486.WEIGL, P. P. 1978. Resource overlap, interspecific interactions and the distribution of theflying squirrel, Glaucomys volans, and G. sabrinus. American Midland Naturalist,100:83-96.WELLS-GOSLING, N., and L. HEANEY. 1984. Glaucomys sabrinus. MammalianSpecies, 229: 1-8.WHITE, T. R. C. 1978. The importance of a relative shortage of food in animal ecology.Oecologia, 33:71-86.WITT, J. W. 1992. Home range and density estimates for northern flying squirrels,Glaucomys sabrinus, in western Oregon. Journal of Mammalogy, 73:921-929.ZAR, J. H. 1984. Biostatistical Analysis. Prentise Hall, New Jersey. 718 pp.61

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0075246/manifest

Comment

Related Items