Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Effects on long-toed salamanders (Ambystoma macrodactylum) of removing canopy cover adjacent to breeding… Ferguson, Christine May 2000

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

Item Metadata

Download

Media
831-ubc_2001-0192.pdf [ 2.64MB ]
Metadata
JSON: 831-1.0090036.json
JSON-LD: 831-1.0090036-ld.json
RDF/XML (Pretty): 831-1.0090036-rdf.xml
RDF/JSON: 831-1.0090036-rdf.json
Turtle: 831-1.0090036-turtle.txt
N-Triples: 831-1.0090036-rdf-ntriples.txt
Original Record: 831-1.0090036-source.json
Full Text
831-1.0090036-fulltext.txt
Citation
831-1.0090036.ris

Full Text

E F F E C T S ON LONG-TOED SALAMANDERS (AMBYSTOMA MACRODACTYLUM) OF REMOVING CANOPY C O V E R A D J A C E N T T O BREEDING SITES AND IN TERRESTRIAL HABITATS by CHRISTINE MAY F E R G U S O N B.Sc , University of Victoria, 1992 A THESIS SUBMITTED IN PARTIAL FULFILMENT O F THE REQUIREMENTS FOR T H E D E G R E E O F MASTER O F SCIENCE in THE FACULTY OF G R A D U A T E STUDIES (f Fa c u1ty< of Forestry) We accept this thesis as conforming to the required standard THE UNIVERSITY O F BRITISH COLUMBIA November 2000 © Christine May Ferguson, 2000 I n p r e s e n t i n g t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f t h e r e q u i r e m e n t s f o r an advanced degree a t t h e U n i v e r s i t y o f B r i t i s h C o l u m b i a , I agree t h a t t h e L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r r e f e r e n c e and s t u d y . I f u r t h e r agree t h a t p e r m i s s i o n f o r e x t e n s i v e c o p y i n g o f t h i s t h e s i s f o r s c h o l a r l y p u r p o s e s may be g r a n t e d b y t h e head o f my department o r by h i s o r h e r r e p r e s e n t a t i v e s . I t i s u n d e r s t o o d t h a t c o p y i n g o r p u b l i c a t i o n o f t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l n o t be a l l o w e d w i t h o u t my w r i t t e n p e r m i s s i o n . Department o f / ^ ^ x ^ g ^ Sc^L£^x_cjt^ The U n i v e r s i t y o f B r i t i s h C o l u m b i a Vancouver, Canada Date / ^ t t < 23 , Z ^ o ; Abstract Knowledge of the effects of logging on amphibians in British Columbia is limited. The long-toed salamander, Ambystoma macrodactylum, is a pond-breeding species with a relatively widespread distribution in the province. I examined the effects of removing canopy cover on long-toed salamanders in both aquatic and terrestrial habitats. The study was conducted at the Opax Mountain Silvicultural Systems Research Area, near Kamloops, B.C. I compared the relative abundance of breeding salamanders in ponds with canopy cover conditions ranging from completely open to natural levels of canopy cover. Similar comparisons were made for juvenile salamanders emerging from breeding ponds. Capture rates of breeding salamanders were positively related to canopy cover index, and were higher in more permanent ponds. The same pattern was observed for emerging juvenile salamanders at one of the two replicate sites, Mud Lake. The relationship was different at the second site, Opax, where capture rates were negatively related to canopy cover index. Emergence of juveniles tended to start earlier at ponds at the Mud Lake site compared to those at Opax. Juvenile salamanders emerged earlier from more open ponds at the Opax site, but not at the Mud Lake site. Differences in effects between sites may result from the shorter growing season at the higher elevation Opax site, which would place constraints on the number of larvae that obtain a large enough size to reach metamorphosis. Higher temperatures in ponds that receive more sunlight would allow faster larval growth, thus increasing the number of emerging juvenile salamanders. The effect of reduced canopy cover on long-toed salamanders in terrestrial habitats was examined by comparing their relative abundance in 1.7-ha patch cut areas with that in uncut forested controls. The experiment was set up in a split-plot design, with volume of downed wood (high or low) as the split-plot factor. Very low capture rates at one of the two replicate sites limited my ability to detect differences between treatments. The pattern observed at the Mud Lake site suggests that activity of long-toed salamanders may be restricted in patch cuts during the summer. Table of Contents Abstract ii Table of Contents iv List of Tables vi List of Figures vii Acknowledgements ix Chapter 1. General Introduction 1 Description of Study Area 3 Life History of the Long-toed Salamander 5 Chapter 2. Effects of Removing Canopy Cover Adjacent to Breeding Sites on Long-toed Salamanders 7 Introduction 7 Methods 11 Use of Ponds by Long-toed Salamanders Across the Research Area 11 Effects of Canopy Reduction on Breeding Long-toed Salamanders 12 Effects of Canopy Reduction on Juvenile Long-toed Salamanders 14 Habitat Sampling 17 Analyses 19 Results 21 Use of ponds by Long-toed Salamanders Across the Research Area 21 Effects of Canopy Reduction on Breeding Long-toed Salamanders 23 Effects of Canopy Reduction on Juvenile Long-toed Salamanders ...25 Habitat 29 Discussion 31 Chapter 3. Effects of Removing Canopy Cover on Long-toed Salamanders in Terrestrial Habitats and the Importance of Downed Wood 35 Introduction 35 Methods 37 Experimental Design 37 Trapping 38 Habitat Sampling 40 Analyses 42 Calculation of capture rates 42 Analysis of treatment effects 42 Habitat Associations 43 iv Results 43 Effects of Treatments on Capture Rates of Long-toed Salamanders 43 Effects of Treatments on Individual Salamanders 55 Habitat 57 Discussion 59 Chapter 4. Conclusions and Recommendations 62 Literature Cited 65 v List of Tables Table 1. Description of ponds sampled for breeding long-toed salamanders. ...13 Table 2. Description of ponds sampled for emerging juvenile long-toed salamanders 16 Table 3. Results of discriminant function analysis to distinguish between adult and juvenile age groups 26 Table 4. Dates of first emergence of juvenile long-toed salamanders from ponds at Mud Lake and Opax 30 Table 5. Habitat attributes measured at terrestrial pitfall grids 41 Table 6. Results of ANOVA comparing harvesting and downed wood treatments for the 1996 summer trapping sessions 45 Table 7. Results of ANOVAs comparing harvesting and downed wood treatments for the 1996 summer trapping sessions, by age group. 47 Table 8. Percent cover of various habitat attributes by site and block: mean (SE); Mud Lake: n = 6 grids; Opax Mountain: n = 6 grids for canopy, and n = 3 for all other attributes 58 vi List of Figures Figure 1. Layout of harvesting treatments at the Opax Mountain research area 4 Figure 2. The life cycle of the long-toed salamander, with timing of events as they are typically observed at the Opax Mountain research area 6 Figure 3. Locations of canopy cover readings around ponds 18 Figure 4. Distribution of pond sizes (m2) used for breeding by long-toed salamanders (light bars) and those with no evidence of breeding activity (dark bars) 22 Figure 5. Distribution of ponds between pond permanence classes for ponds used for breeding by long-toed salamanders (light bars) and for ponds with no evidence of breeding activity (dark bars) 22 Figure 6. Mean capture rates of adult long-toed salamanders (LTS) in breeding ponds, by site. Error bars are 1 S E 24 Figure 7. Mean capture rates of juvenile long-toed salamanders (LTS) emerging from breeding ponds, by site. Error bars are 1 S E 28 Figure 8. Mean capture rates of long-toed salamanders (LTS) for the 1996 summer trapping sessions in forest and patch cut treatment units by site. Error bars are 1 S E 44 Figure 9. Mean capture rates of long-toed salamanders (LTS) for the 1996 summer sessions in forest and patch cut treatment units at the Mud Lake site, by age group. Error bars are 1 S E 48 Figure 10. Mean capture rates of long-toed salamanders (LTS) for the 1996 summer sessions in high and low downed wood treatment units at Mud Lake by harvesting unit, for a) adult salamanders and b) juvenile salamanders. Error bars are 1 S E 49 Figure 11. Mean capture rates of long-toed salamanders (LTS) for the fall trapping sessions during 1996 in forest and patch cut treatment units by site, for a) adult salamanders and b) juvenile salamanders. Error bars are 1 SE.50 Figure 12. Mean capture rates of long-toed salamanders (LTS) for fall sessions during 1996 in high and low downed wood treatment units by harvesting unit, for a) adult salamanders and b) juvenile salamanders. Error bars are 1 S E 51 vii Figure 13. Mean capture rates of juvenile long-toed salamanders (LTS) at downed wood arrays at Mud Lake in 1995, by harvesting treatment ...53 Figure 14. Change in mean capture rates between summer and fall trapping sessions during by harvesting treatment, for a) adult long-toed salamanders (LTS) and b) juvenile salamanders. Error bars are 1 S E 54 Figure 15. Comparison of mean capture rates at the Mud Lake and Opax sites by season, for a) adult long-toed salamanders (LTS) in 1996, b) juvenile salamanders in 1996, and c) juvenile salamanders in 1995. Error bars are 1 S E 56 viii Acknowledgements I thank my committee, Tom Sullivan, Michael Feller and Alton Harestad, for their time.and input, and Walt Klenner for his input and assistance in the initial stages of my project. Thanks to my field assistant, Sara Short, for her hard work, enthusiasm, and support, and to Darren Ferguson, Shane Vermeulen, Ralph Heinrich, Tamsin Baker, and Gillian Turney for help with installing and/or checking traps, would also like to thank my friends and family, who have been helpful and supportive throughout the various stages of this thesis. Special thanks to Vanessa Craig, Dave Huggard and Alan Vyse for their help, support and encouragement. Funding for this project was provided by Forest Renewal BC (project no. TO96072-RE). Chapter 1. General Introduction The importance of natural biodiversity in forests has received increased recognition in recent years, as reflected by the inclusion of biodiversity management guidelines in the Forest Practices Code of British Columbia. Amphibians are an important, though somewhat invisible component of forest biodiversity, differing considerably from other vertebrates in their physiology, behaviour and role in food webs. Their ability to convert energy efficiently makes them an important food link between larger vertebrates and the biomass of small invertebrates (Pough 1983). The effects of habitat alteration on wildlife species are often variable. Some species may benefit from changes, while others will be adversely affected. Aspects of amphibian biology (e.g. permeable skin, ectothermy) restrict their ability to deal with fluctuating conditions, making them potentially sensitive to habitat alteration. Loss of habitat has been identified as a key factor contributing to declines in amphibian populations, detected in recent years around the world (Blaustein er al. 1994). Reductions in species richness, diversity and the overall abundance of amphibians have been associated with logging in several forest ecosystems. These include forests in coastal British Columbia (Dupuis etal. 1995), the Pacific Northwest (Raphael 1988; Welsh and Lind 1988) and the eastern United States (Enge and Marion 1986; Pough etal. 1987; Petranka etal. 1994). Forest harvesting can affect amphibians in aquatic habitats as well (Corn and Bury 1989). Most studies of the effects of forestry practices on amphibians have been in the United States, and very few examine effects in both terrestrial and aquatic 1 environments. It is difficult to extrapolate these results to ecosystems in the interior of British Columbia, due to differences in climate and in the amphibian species present. There is little information on the terrestrial habitats used by the amphibian species resident in BC, and even less on the effects of logging on these species (Davis 1999). The long-toed salamander, Ambystoma macrodactylum, is the most widespread species of salamander in British Columbia, and the province represents a substantial portion of its total distribution. Although some North American amphibians are entirely terrestrial or entirely aquatic, most depend on both terrestrial and aquatic habitats to complete their life cycle. The species found in the interior of British Columbia typically breed in temporary or permanent ponds in the spring, and move into the terrestrial environment for the remainder of the year. For long-toed salamanders, the onset of the breeding season varies with elevation and latitude, largely in response to climatic factors (Ferguson 1961). Coastal and low elevation populations are thought to be most influenced by rainfall, while populations in locations with cold winters respond primarily to temperature (Anderson 1967). Aquatic habitats must provide conditions suitable for the development and growth of egg and larval stages (e.g. sufficient food and cover; suitable temperatures). Terrestrial habitats must provide cover and food for post-metamorphic life stages (Orchard 1988). Downed wood may play an important role in providing protection from desiccation and extremes of 2 temperature in the terrestrial environment, as well as providing an abundance of small invertebrates for food. Our understanding of the impacts of forest management on amphibians in British Columbia is limited, especially for the dry interior of the province. Information is required to evaluate the effectiveness of current Riparian Management and Biodiversity guidelines in the Forest Practices Code for maintaining amphibians in forest ecosystems. The aim of this study was to improve our understanding of the effects of forest harvesting on long-toed salamanders in Interior Douglas-fir forests, in both terrestrial and aquatic habitats. In Chapter 2, I outline characteristics of ponds used by long-toed salamanders, and examine the effects of reducing the canopy cover, adjacent to breeding ponds, on the breeding population size and abundance of emerging juvenile salamanders. In Chapter 3, I examine the importance of canopy cover and downed wood to long-toed salamanders. General conclusions and recommendations are presented in Chapter 4. Description of Study Area This study was conducted at the Opax Mountain Silvicultural Systems research area, located approximately 20 km northwest of Kamloops, B.C., in the IDF biogeoclimatic zone. The research area was set up to examine alternative approaches to managing dry Douglas-fir (Pseudotsuga menziesii) forests, which have traditionally been managed by stand-level partial cutting (Klenner and Vyse 1998). 3 The research area is comprised of two sites. The Mud lake site, at elevations between 950 and 1100 m, is in the IDFxh2 subzone. The Opax site is classified as IDFdkl, and ranges in elevation from 1200 to 1370 m. Harvesting treatments were applied to the six 20-ha treatment units at each site during the winter of 1993-1994. These were 50% removal by uniform partial cutting, 50% removal in variable-sized patch cuts, 20% removal by uniform partial cutting, 20% removal in variable-sized patch cuts, 35% removal (50% uniform removal with uncut reserve areas), and a control (no harvesting) (Figure 1). More than 100 wetlands of varying size and type are found across the research area. These include marshes, small open water ponds and one lake (at the centre of the Mud Lake site). Figure 1. Layout of harvesting treatments at the Opax Mountain research area. 4 Life History of the Long-toed Salamander The long-toed salamander is an aquatic-breeding amphibian with an estimated life expectancy of six years (Russell et al. 1996). Sexual maturity is reached in two to three years (Green and Campbell 1984). Adults are primarily terrestrial, returning to breeding sites each year to reproduce. The timing of events in the life-cycle of this salamander varies across its range and with elevation, with breeding occurring in winter in milder portions of its range and as late as midsummer in areas with more extreme winter conditions (Howard and Wallace 1985). At the Opax Mountain research area, breeding activity typically begins late March into April, depending on the timing of ice-melt at breeding sites, and continues through May (Figure 2). Eggs are deposited on aquatic vegetation or other subsurface objects (Orchard 1988). The carnivorous larvae typically metamorphose in their first season, although they may overwinter at higher elevations (Howard and Wallace 1985). Emergence of metamorphs begins early July and continues into late October at this site. Long-toed salamanders overwinter in terrestrial habitats (Orchard 1984). 5 E g g s deve lop L a r v a e deve lop Breeding Overwinter ing E m e r g e n c e o f . . . metamorphs Figure 2. The life cycle of the long-toed salamander, with timing of events as they are typically observed at the Opax Mountain research area. 6 Chapter 2. Effects of Removing Canopy Cover Adjacent to Breeding Sites on Long-toed Salamanders Introduction Habitat loss is one of several factors cited as contributing to declines in amphibian populations in North America (Blaustein and Wake 1990). The loss of wetlands through drainage, contamination, and development is especially relevant for pond-breeding species of amphibians. The extent to which forestry activities affect wetland habitats is not well known, although studies have shown that logging adjacent to streams impacts both fish and amphibians. Corn and Bury (1989) found that the species richness of stream-dwelling amphibians, as well as density and biomass, was lower for streams that had logging in their headwaters. The Riparian Management Guidelines, as part of the Forest Practices Code of British Columbia, are intended to maintain adequate water quality and wildlife habitat (B.C. Ministry of Forests and B.C. Environment 1995). Riparian Management Areas (RMAs) consisting of a reserve and/or management zone are recommended for streams, lakes and wetlands of particular classifications. The classification scheme for streams depends on size, whether or not a stream is in a community watershed, and on the presence offish. For lakes and wetlands, classification is based on size and biogeoclimatic unit. To date, there have been no studies that examine the effectiveness of these guidelines for protecting amphibians (Davis 1999). Within most biogeoclimatic zones, no RMA is required for wetlands smaller than 1 ha; in selected subzones where wetlands 7 are less common, RMAs are required for wetlands as small as 0.5 or 0.25 ha. Smaller wetlands, which may often be ephemeral, provide preferred breeding habitats for many amphibian species (Walls er al. 1992), and may be important to the persistence of amphibian populations across large spatial scales (Gibbs 1993; Richter and Azous 1995). The long-toed salamander, Ambystoma macrodactylum, breeds in ponds, lakes and slow-moving streams (Orchard and Harcombe 1988), depositing eggs on aquatic vegetation or other subsurface objects (Orchard 1988). Use of aquatic habitats by long-toed salamanders is generally limited to breeding activity and the development of egg and larval stages, as adults are primarily terrestrial outside of the breeding season. Studies in natural and artificial ponds have identified several factors that influence the growth, development and survivorship of egg and larval stages. These include physical factors, such as temperature, pH, UV radiation and pond permanence, as well as biological factors, including food availabiJity, larval density and the presence of predators. Effects experienced in the early life stages of amphibians (e.g., effects on size and condition, reductions in numbers) may be carried over into adult stages, thus affecting the future population of breeding individuals (Berven 1990; Scott 1994). Within the biologically tolerable range of a species, higher temperatures can increase growth and development rates. For larval Ambystoma tigrinum, growth rates and size at metamorphosis were positively correlated with temperature (Petranka 1984). In a laboratory study of larval Hyla gratiosa and H. cinerea reared at two different temperatures, the larval period was consistently 8 shorter at the higher temperature (Leips and Travis 1994). The hatching success of eggs at breeding sites with low pH may be reduced (Clark 1986), although tolerance for low pH conditions varies among species (Dale et al. 1985; Freda and Dunson 1986). Blaustein er al. (1994) found that exposure to UV-B radiation reduced the hatching success of eggs of some amphibian species (Rana cascadae, Bufo boreas), while others were not affected (Hyla regilla). Pond permanence can influence species composition and the abundance of individual species. Ephemeral ponds are preferred by some species due to their high productivity and the exclusion of some predators (Walls et al. 1992), while others require permanent water because their larvae do not metamorphose in their first season. Pond duration influenced yearly variation in the survival of juvenile Ambystoma maculatum (Shoop 1974), and Pechmann et al. (1989) found that both species diversity and the number of juveniles metamorphosing in three ephemeral ponds were higher in years the ponds held water for a longer period. Decreasing water levels resulted in faster development of larvae and a smaller size at metamorphosis in Hyla pseudopuma (Crump 1989) and Ambystoma talpoideum (Semlitsch and Wilbur 1988). Pond permanence was a factor in the distribution of Pseudacris treefrog larvae (Skelly 1995). Faster growth rates of amphibian larvae have been associated with higher food availability for Ambystoma texanum (Petranka 1984), A. gracile (Licht 1992) and Scaphiopus couchii (Newman 1994). Larvae of Hyla gratiosa and H. cinerea had shorter larval periods and larger sizes at metamorphosis when raised at higher food levels (Leips and Travis 1994). Reduced food availability is one 9 effect of higher larval densities, but growth may also be affected by density independently of food availability. At equal per capita food availability, Scaphiopus couchii tadpoles grew more quickly at lower tadpole densities (Newman 1994). Scott (1994) found that larval density affected size and lipid stores in both juvenile and adult stages Ambystoma opacum, as well as time to first reproduction and size at reproduction. The survival and size of larval A. opacum was increased at lower densities (Petranka 1989), and larval density was a good predictor of growth rate, size at metamorphosis and survivorship to metamorphosis for Ambystoma texanum (Petranka and Sih 1986). The presence of predators can have a significant effect on survival from egg to adult stages (Licht 1974), and can reduce larval growth rates by limiting larval activity (Holomuzki 1986; Skelly 1995). The forest canopy adjacent to a pond influences the amount of light penetration, potentially affecting several of the factors mentioned above. As a result, changes in adjacent canopy cover have the potential to influence resident amphibian communities. A negative impact may be expressed as a reduction in numbers of juveniles emerging from ponds and/or poorer condition of individuals. Over time this would result in a smaller breeding population, because many amphibians show high site fidelity. My study was designed to test the hypothesis that increasing levels of canopy removal adjacent to ponds would 1) reduce the relative abundance of breeding long-toed salamanders, 2) reduce the number of juvenile long-toed salamanders emerging from ponds, and 3) negatively affect the size and condition of individual salamanders. 10 Methods Use of Ponds by Long-toed Salamanders Across the Research Area Before selecting the ponds I sampled to test the above hypotheses, I wanted to determine the general characteristics of ponds that represented suitable breeding habitat. I surveyed the research area in 1995, locating 112 ponds, and assessed the size and permanence of each pond. I measured the length (maximum dimension) and width (dimension perpendicular to the length) of each pond and used these to calculate an index of surface area (size = 0.8 * length * width). As most ponds were relatively symmetrical, this gave a reasonable estimate of surface area. For ponds with irregular shapes, the size index was determined by splitting the pond into sections and combining the indices for each. To determine the permanence and drying rates of ponds, wooden stakes with markings at increments of 1 cm were deposited in each pond, and the water level monitored over time. Water depths were monitored from early June to October and the date when each pond dried up in 1995 and in 1996 was noted. Ponds were assigned to pond permanence classes based on their dry dates: permanent ponds were those that retained water beyond mid-September in both years, temporary ponds dried before the end of July in both years, and intermediate ponds were those that dried after July but before mid-September in at least one year. 11 I used funnel traps made from 2-L plastic pop-bottles (Adams et al. 1997) to capture long-toed salamanders breeding in ponds during spring of 1995. Sampling effort at any one pond was low, so that many ponds across the research area could be sampled. Searches for eggs and larvae later in the season provided further indication of which ponds were used for breeding. Effects of Canopy Reduction on Breeding Long-toed Salamanders Application of the partial cutting treatments across the research area resulted in ponds with a range of adjacent canopy cover conditions. I measured the relative abundance of breeding salamanders at ponds with different levels of canopy removal, at two replicate sites, Mud Lake and Opax. I grouped ponds into two broad canopy cover categories (low/no canopy removal; moderate to high canopy removal) and selected seven ponds from each category at each site. The 14 ponds at each site had canopy cover conditions ranging from complete canopy removal to no canopy removal (including natural variation in canopy cover). Ponds that dried before the end of May in 1995 were excluded from this process. Table 1 lists the size index, permanence category and index of adjacent canopy cover for ponds sampled. Each of the 28 ponds was sampled over three consecutive nights, with traps checked each morning. Because breeding activity was predominantly nocturnal, traps could be left set in the pond after each morning check without salamanders spending too much time in traps. Sampling occurred during two trapping sessions: 18 ponds were sampled between April 13 and May 2 and 10 12 Table 1. Description of ponds sampled for breeding long-toed salamanders. Site Pond No. Size Index (m2) Pond Permanence Canopy Cover Index Mud Lake 1 276 PERM 14.96 19 96 INT 4.66 23 a 720 PERM 19.07 26 211 INT 13.51 27 237 INT 6.57 32 168 TEMP 4.21 322 40 TEMP 10.38 •34 269 TEMP 8.29 35 a 83 PERM 2.18 39 101 TEMP 14.96 40 a 202 PERM 18.31 42 a 320 PERM 6.04 44 a 528 PERM 15.96 49 250 TEMP 6.84 Opax 52 a 528 PERM 12.76 53 1931 PERM 20.87 54 a 168 PERM 15.68 58 136 INT 2.51 62 a 112 INT 11.38 66 106 TEMP 13.03 67 a 168 PERM 8.41 68 a 608 PERM 0.00 75 a 516 PERM 0.43 77 a 36 INT 0.00 78 a 96 INT 5.56 80 a 474 INT 6.82 93 34 TEMP 0.47 94 72 TEMP 13.31 Pond Permanence: PERM- permanent, INT- intermediate, TEMP- temporary. Canopy Cover Index: maximum value possible is 25. a These ponds were sampled for both adults and emerging juveniles. 13 were sampled between May 8 and May 24. Within each session, the Mud Lake ponds were sampled before the Opax ponds, reflecting the later ice melt at Opax. One pond at the Opax site, pond 54, was unique in that it had very little melting by early May, prohibiting entry of breeding salamanders. Most other ponds had at least partially melted at this time, although they often had a thin layer of ice overnight. Because the designation of trapping session was included to account for sampling later in the breeding season, pond 54 was considered to be from session one for the analysis. Breeding long-toed salamanders were live-trapped in ponds using pop-bottle funnel traps placed 1 m from the pond edge, at intervals of 5 m. I trapped an air bubble in each funnel trap to prevent oxygen deficiencies in the trap when capture rates were high. The snout-to-vent length (SVL), total length, weight and sex of each salamander was recorded. SVL was measured from the tip of the snout to the anterior end of the vent opening (approximated as the point where the posterior of the back legs connect). Weight, to the nearest 0.1 g, was measured with a Pesola scale. Tail damage including the loss of a portion of the tail was often observed, and was noted. Effects of Canopy Reduction on Juvenile Long-toed Salamanders Not all ponds sampled in the spring breeding experiment could be sampled for emerging juvenile salamanders. All temporary ponds were excluded from the second experiment, as these were dry prior to installation of drift fences. Other ponds were excluded because they were difficult to sample with drift 14 fences, due to either their large size or their proximity to other ponds. This reduced the number of ponds sampled at the Opax site to nine. Twelve ponds were sampled at the Mud Lake site, including five of the original ponds and seven new ones (Table 2). Sampling commenced in early August and continued into mid-October, providing 70 and 59 consecutive days of sampling at Mud Lake and Opax, respectively. In addition to a measurement of relative abundance, this provided information on the timing of emergence of juvenile salamanders and the size of juveniles at emergence. Juvenile long-toed salamanders were captured as they emerged from breeding ponds, in pitfall traps set adjacent to 5-m long drift fences. Three pitfall traps were installed along each fence, one in the middle and one at each end. Fences were constructed of 30-cm wide strips of heavy plastic, with an above-ground height of 20 cm and 10 cm buried below ground. Pitfall traps were 12.5 cm deep, with a funnel rim to prevent salamanders from crawling out. Wet moss or moist decayed wood was put in the bottom for cover from predators and to prevent desiccation. I installed a minimum of three drift fences at each pond, one for every 20 m of perimeter (or at equal spacing for ponds with a perimeter less than 60 m). Fences followed the contour of the pond at a distance of 2 m from the edge. I measured snout-to-vent length (SVL), total length and weight, and noted the condition of emerging juvenile salamanders (injuries, deformities, etc.). When capture rates were high, I recorded lengths and weights for every fourth 15 Table 2. Description of ponds sampled for emerging juvenile long-toed salamanders. Site Pond no. Size Index Pond Canopy Cover (m2) Permanence Index Mud Lake 5 392 PERM 13.16 6 259 INT 18.82 7 58 INT 13.04 23 a 720 PERM 19.07 28 120 INT 18.34 31 65 INT 1.41 35 a 83 PERM 2.18 40 a 202 PERM 18.31 42 a 320 PERM 6.04 43 84 INT 4.18 44 a 528 PERM 15.96 101 590 PERM 6.03 Opax 52 a 528 PERM 12.76 54 a 168 PERM 15.68 62 a 112 INT 11.38 67 a 168 PERM 8.41 68 a 608 PERM 0.00 75 a 516 PERM 0.43 77 a 36 INT 0.00 78 a 96 INT 5.56 80 a 474 INT 6.82 Pond Permanence: PERM- permanent, INT- intermediate, TEMP- temporary. Canopy Cover Index: maximum value possible is 25. a These ponds were sampled for both adults and emerging juveniles. 16 salamander removed from the trap. Animals were released nearby, on the opposite side of the fence. In addition to the emerging cohort of salamanders, adult and yearling salamanders were captured at fences. In most cases juvenile salamanders could be identified by their small size, the presence of gill marks and their less developed dorsal colour pattern, but in some cases it was difficult to distinguish between newly emerged salamanders and young salamanders from the previous year. I estimated the age group of these from their weights and measurements (see analysis section below). Habitat Sampling The following habitat variables were measured at all ponds sampled for breeding and/or emerging juvenile long-toed salamanders: 1) Adjacent canopy cover: Vertical canopy readings were taken using a "moosehorn" device, at eight locations around each pond (N, NE, E, etc.), with three distances for each location (at the pond edge, and 5 m and 10 m out from the edge) (Figure 3). An index of adjacent canopy cover was calculated by weighting the readings so as to emphasize the canopy cover readings that would have the most influence on the amount of sunlight reaching the pond over a day. Readings taken E and W of each pond were weighted by a factor of three, S E and SW by a factor of two, and northerly readings (N, NE, NW) were excluded from the index. An 17 10 10 c 10 • 5 • 5 * 5 Figure 3. Locations of canopy cover readings around ponds. 18 additional weighting was applied for distance from pond edge (three for pond edge, two for 5 m, and one for 10 m). 2) Percent cover of aquatic vegetation: Estimates of the percent cover at the pond surface of emergent and of floating vegetation were recorded in six cover classes (<5%; 5.1-25%; 25.1-50%; 50.1-75%; 75.1-95%; >95%). 3) Percent cover of woody debris: Percent cover was recorded for woody debris in each pond in two size classes (diameter: < 12 cm, >= 12 cm) using the six cover classes described above. 4) Substrate type: The pond substrate was described as either solid or soft and the substrate/litter types present were recorded (e.g., rock, mud, deciduous leaves, etc.). 5) pH: pH was measured at five locations per pond with a Piccolo 2 A T C pH-meter. Analyses I used a Kolmogorov-Smirnov two-sample test to compare the size distribution of ponds with evidence of breeding activity by long-toed salamanders to the size distribution of ponds without evidence of breeding. Ponds that were not adequately checked for evidence of breeding activity were excluded from the analysis. Size classes were < 100 m 2 , 100.1-200 m 2 , 200.1-300 m 2 , 300.1-500 m 2 , 500.1-1000 m 2 , and > 1000 m 2. A similar analysis was used to compare the permanence (permanent, intermediate, temporary) of ponds used for breeding by long-toed salamanders with those not used. 19 I used t-tests to test for differences in the mean capture rates of breeding long-toed salamanders for ponds at Mud Lake versus Opax, and for differences in the mean capture rates of emerging juvenile salamanders. Welch's approximate t was used to compare mean capture rates of juvenile salamanders between sites, because this test does not assume equal variances. I used stepwise multiple regression to determine which independent variables explained variation in the capture rates of both breeding and juvenile long-toed salamanders between ponds, and variation in the median emergence dates of juvenile salamanders. Site, canopy cover index, pond permanence, pond size and trapping session were the variables included in the regression analysis for breeding salamanders. Analyses for juvenile salamanders included canopy cover index, pond permanence, and pond size, with separate analyses for each site. Correlation analysis was used to test for colinearity between variables. Categorical variables were included in the above analyses as dummy variables. Alpha was set at 0.05 for all tests. I used discriminant function analysis to assign salamanders of uncertain age into two age groups: juveniles (emerging juveniles of the 1996 cohort) and adults (salamanders that emerged prior to 1996). The analysis was based on measurements of SVL, total length and weight for 638 salamanders of known age group. 20 Results Use of ponds by Long-toed Salamanders Across the Research Area All but two of the 112 ponds within the research area had a surface area of less than 0.25 ha. The remaining two ponds had size indices of 17600 and 7360 m 2 . Ninety percent of ponds had a size index of less than 800 m 2 and the median value was 129 m 2 . Ponds sampled for the spring breeding experiment ranged in size from 34 to 1931 m 2 (Table 1), and ponds for the juvenile emergence experiment ranged in size from 58 to 720 m 2 (Table 2). Results of the Kolmogorov-Smirnov test did not indicate differences between the size distributions of ponds used for breeding and those not used for breeding by long-toed salamanders. Ponds smaller than 100 m 2 were less common in the sample of ponds used for breeding, but this difference was not significant (p > 0.05) (Figure 4). Significant differences were found between pond permanence distributions for ponds with and without evidence of breeding (p < 0.05) (Figure 5). Temporary ponds were most abundant on the site, but were less frequently used for breeding than the other classes of ponds. Less than 40% of adequately checked temporary ponds had evidence of use for breeding, compared to 83% for permanent ponds and 66% for intermediate ponds. In addition to the 11 temporary ponds used for breeding, long-toed salamanders were observed in five other temporary ponds, although there was no sign that breeding had occurred at these sites. The earliest drying pond with evidence of breeding 21 0.50 T <100 100.1- 200.1- 300.1- 400.1- 500.1- >1000 200 300 400 500 1000 Size class Figure 4. Distribution of pond sizes (m2) used for breeding by long-toed salamanders (light bars, n = 57) and those with no evidence of breeding activity (dark bars, n = 31). 0.60 0.50 c 0.40 o o 0.30 CL O £ 0.20 0.10 0.00 Permanent Intermediate Temporary Pond permanence Figure 5. Distribution of ponds between pond permanence classes for ponds used for breeding by long-toed salamanders (light bars, n = 57) and for ponds with no evidence of breeding activity (dark bars, n = 31). 22 activity was dry by June 21 in 1995 and July 8 in 1996; desiccated salamander larvae were found in the dry bottom of this pond. Similar proportions of permanent and intermediate ponds were used by salamanders for breeding (Figure 5). Pond size and pond permanence showed a moderate association (n = 94, r = 0.295, p = 0.004). Temporary ponds tended to be small, and were never more than 400-m 2 in size. Most intermediate ponds had size indices less than 1000 m 2 . Permanent ponds exhibited a full range of pond sizes, and 90% of ponds larger than 1000 m 2 were permanent. Effects of Canopy Reduction on Breeding Long-toed Salamanders Estimates of relative abundance are based on 1051 captures of breeding long-toed salamanders, 686 at Mud Lake and 365 at Opax Mountain. Overall capture rates for ponds were not significantly different between the two sites (t = 0.49, df = 26, p = 0.63), with mean values of 51 and 38 salamanders per 100 trap-nights for Mud Lake and Opax Mountain, respectively (Figure 6). Canopy cover index, pond permanence and trapping session explained a significant portion of the observed variation in capture rates of breeding long-toed salamanders between ponds (n = 28 ponds, R 2 = 0.72, F = 8.91, p < 0.001). The regression equation was: Y = 0.003 (CANCOV) 2 + 0.504 (DRY1) + 0.854 (DRY2) - 0.873 (TRAPSESSION) where DRY 1 and DRY2 are the dummy variables representing pond permanence. Capture rates for breeding salamanders were higher for more 23 100 f) 80 'E ci g 60 CL CO 40 20 1 1 1 Mud Lake Opax Site Figure 6. Mean capture rates of adult long-toed salamanders (LTS) in breeding ponds, by site. Error bars are 1 S E . 24 permanent ponds, and they were higher in the first trapping session. Capture rates showed a positive association with canopy cover index. Site and pond size were not significant variables in the regression analysis. Correlation analysis indicated that canopy cover index and pond size were positively associated (n = 28, r = 0.395, p = 0.04), but closer observation showed this to result from the influence of one point. No association between variables was apparent following removal of this pond from the analysis (n = 27, r = 0.166, p = 0.41). There was no association between canopy cover index and pond permanence (n = 28, r = 0.186, p = 0.34). Effects of Canopy Reduction on Juvenile Long-toed Salamanders Captures at drift fences were not limited to recently emerged salamanders (the 1996 cohort). Salamanders of uncertain age group were assigned to a group based on the results of the discriminant function analysis before running the analysis of capture rates of emerging juveniles. The discriminant function was very successful at discriminating between adult and juvenile salamanders of known age group, assigning individuals to the correct group 99% of the time (Table 3). a) Relative Abundance Estimates of relative abundance are based on 2736 captures of juvenile long-toed salamanders emerging from breeding ponds, 2582 at Mud Lake and 154 at Opax. Capture rates for ponds at the Mud Lake site were significantly 25 Table 3. Results of discriminant function analysis to distinguish between adult and juvenile age groups. Canonical discriminant functions standardized by within variances Snout-vent length 0.274 Total length 0.141 Weight 0.661 Classification Matrix Juveniles Adults % Correctly Classified Juveniles 510 1 100 Adults 5 122 96 Total 515 123 99 26 higher than at the Opax site (Welch's t = 3.78, df = 11, p = 0.003). The mean capture rate at Mud Lake was 63 salamanders per 100 fence-nights versus 7 salamanders per 100 fence-nights at Opax (Figure 7). Observed trends differed between sites as well. Both canopy cover index and pond permanence were significant variables at the Mud Lake site (n = 12 ponds, R 2 = 0.81, F = 20.90, p < 0.001). The regression equation was: Y = 0.212 (CANCOV) 2 + 40.279 (DRY) where DRY is the dummy variable representing more permanent ponds. The relationship between canopy cover index and capture rate was positive, and higher capture rates of emerging juvenile salamanders were associated with the more permanent ponds. At Opax, however, canopy cover index was the only significant variable and it showed a negative relationship with juvenile capture rate (n = 9 ponds, R 2 = 0.59, F = 9.99, p = 0.02). The equation was: Y = - 0.840 (CANCOV) + 12.945 There was no association between canopy cover index and pond size (n = 21, r = 0.113, p = 0.63) or canopy cover index and pond permanence (n = 21, r = 0.075, p = 0.75). b) Timing of Emergence At Mud Lake, variation in median emergence date between ponds was explained by both canopy cover index and pond permanence (n = 12 ponds, R 2 = 0.65, F = 4.86, p = 0.03). The regression equation was: Y = 2.850 (CANCOV) - 0.132 (CANCOV) 2 - 10.652 (DRY) + 23.905 27 c 'E d> o c 0) 80 70 60 50 -f 40 30 4-§_ 20 OT 10 J 0 Mud Lake Opax Site Figure 7. Mean capture rates of juvenile long-toed salamanders (LTS) emerging from breeding ponds, by site. Error bars are 1 S E . 28 Juvenile salamanders emerged earlier from permanent ponds. Emergence was later at ponds with intermediate values of canopy cover index. Median emergence dates for ponds at the Opax site showed a positive relationship with canopy cover index (n = 9 ponds, R 2 = 0.65, F = 13.044, p = 0.009); juvenile salamanders emerged earlier from ponds with a lower canopy cover index. The equation was: Y = 0.128 (CANCOV) 2 + 20.643 There was less variation in median emergence dates at Mud Lake than at Opax. The range at Mud Lake was August 27 to September 23 versus August 22 to October 3 at Opax. Emergence of juvenile salamanders tended to start earlier from ponds at the Mud Lake site compared to the Opax site (Table 4). By August 23, larvae had started to emerge at all Mud Lake ponds, but only at three ponds at Opax. Emergence at Opax was earliest at pond 77, which was also the first pond to dry. Habitat The range of pH values at the Mud Lake site was 7.2 to 8.9, somewhat higher than for ponds at the Opax site, where the range was 6.1 to 7.3. Pond depths were monitored in 1995 and 1996. Overall, 1995 was a drier year than 1996, and most of the intermediate ponds dried earlier in 1995 than 1996. Water levels rose in mid-August in 1995, compared to early September in 1996. Ponds that were not dry by these dates tended not to dry out in that year, 29 Table 4. Dates of first emergence of juvenile long-toed salamanders from ponds at Mud Lake and Opax. Mud Lake Earliest date of Opax Earliest date of ponds emergence ponds emergence 43 July 23* 77 July 24* 40 July 24* 78 August 11 44 July 24* 75 August 22 23 August 8 68 August 26 31 August 8 80 August 26 35 August 12 52 September 2 6 August 14 54 September 2 28 August 14 67 September 2 42 August 16 62 September 26 5 August 23 7 August 23 101 August 23 *ln a few cases juvenile salamanders were captured emerging from ponds immediately following installation of drift fences. The date of first emergence for these ponds may be earlier than stated. 30 so that whether or not an intermediate pond dried in a year depended on the timing of late summer rains. Discussion Long-toed salamanders did not show a preference for breeding sites of a particular size, breeding in ponds smaller than 100 m 2 as well as in the large lake. The large number of salamanders breeding in, and emerging from, small ponds suggests that these ponds are an important habitat component in this forest type. Richter and Azous (1995) found that smaller wetlands in the Puget Sound Basin offered "high-value" amphibian habitat compared to larger ones. \ Most ponds across the Opax Mountain research area were small, and only two ponds on the entire site would require a Riparian Management Area under the guidelines in the Riparian Management Guidebook of the Forest Practices Code. One would be classed as a W2 wetland (10 m reserve zone; 20 m management zone) and the other as a W4 wetland (no reserve zone; 30 m management zone). The remaining 110 ponds are unclassified wetlands, and therefore require no management zone. If logging activities adjacent to breeding sites negatively impact salamander populations, the loss of these smaller wetlands would reduce the availability of suitable breeding sites. The loss of small wetlands as suitable habitat can greatly increase inter-wetland distances, affecting dispersal and population recovery following local extirpations (Gibbs 1993). The positive relationship between capture rates of breeding long-toed salamanders and canopy cover index suggests that population sizes may be 31 smaller at ponds with lower canopy cover. Higher rates of juvenile emergence at the Mud Lake site were associated with a higher canopy cover index as well. Increased exposure to UV-radiation following reductions in canopy cover is one of several possible mechanisms leading to reduced survival of aquatic life stages. Blaustein et al. (1994) found that embryos of different species of amphibians were affected to different degrees by UV-B radiation. These authors related the differences to levels of photolyase, a protein involved in the repair of DNA. Levels of photolyase appear correlated to the degree of concealment or depth of placement of eggs, and salamanders generally have lower photolyase activities than anurans. Eggs laid in shallow water might be protected by the canopy adjacent to breeding ponds. Removal of the canopy may expose eggs to levels of UV radiation higher than can be tolerated. The negative relationship between juvenile emergence rates and canopy cover index at the Opax site contradicts the trend described above. The generally low rate of emergence at Opax compared to Mud Lake, despite breeding capture rates of similar magnitude between sites, suggests that survival to metamorphosis may be more constrained at Opax. The onset of the breeding season for long-toed salamanders varies with elevation and latitude, largely in response to climatic factors (Ferguson 1961). Coastal and low elevation populations are thought to be most influenced by rainfall, while populations in locations with cold winters respond primarily to temperature (Anderson 1967). At the Opax Mountain research area, the onset of breeding coincides with the melting of ice on the ponds in April and May. This appeared to start at least two 32 weeks later at Opax ponds compared to Mud Lake ponds, especially for those with high canopy cover. A similar difference was noted for the timing of snow melt between the two sites on the research area (Huggard et al. 1998). Amphibian larvae must reach a minimum size before they can undergo metamorphosis (Wilbur 1980), but if this size is not reached in the first year, then larvae may overwinter if conditions permit. The overwintering of long-toed salamander larvae has been observed more commonly at higher elevations (Howard and Wallace 1985). The shorter growing season at Opax may restrict the number of larvae that reach the minimum size required for metamorphosis in their first season. If this is the case, higher water temperatures following reductions in canopy cover may be leading to higher growth rates for more open ponds, thereby increasing abundance. Another expected result of increased growth rates would be earlier emergence of juvenile salamanders, as larvae would reach the minimum size required for metamorphosis more quickly. Juvenile salamanders emerged earlier from ponds with lower canopy cover at the Opax site, again suggesting that growth rates may have been increased by elevated water temperatures at low canopy cover ponds. Higher abundances of breeding long-toed salamanders were associated with more permanent ponds, and permanent ponds had the highest proportional use compared to intermediate and temporary ponds. Pechmann et al. (1989) found that emergence rates for juvenile Ambystoma talpoideum were strongly related to the number of days that ponds held water, although this was not the 33 case for other amphibian species at the same sites. The breeding population size in these more predictable sites may be larger because juvenile recruitment is more reliable. Higher rates of juvenile emergence were found for permanent ponds compared to intermediate ponds at Mud Lake, but not at Opax Mountain. If the growing season is more restricted at Opax as I suggested above, this effect may have been overshadowed by the strong effect of canopy cover index on juvenile emergence rates. 34 Chapter 3. Effects of Removing Canopy Cover on Long-toed Salamanders in Terrestrial Habitats and the Importance of Downed Wood Introduction Forest ecosystems undergo many physical and biological changes through forest management. Loss of the forest canopy results in greater fluctuations in air and soil temperatures, increased wind speed, and changes in the moisture regime (Chen et al. 1993). This change in the physical environment can lead to changes in biological species composition, as some species may be eliminated or reduced in density while others thrive under the new conditions. Amphibians are considered to be quite susceptible to these kinds of changes, because aspects of their biology (e.g., ectothermy; permeable skin) restrict their ability to deal with fluctuating conditions. Behavioural responses, such as limiting activity to cooler seasons or times of day, are important in the avoidance of harsh conditions. Many amphibians use downed wood as cover, and some terrestrial breeding species lay their eggs in large decayed logs. Forest management can result in a reduction in coarse woody debris (CWD) input into a stand over time. There are concerns that even as stands mature, input rates of downed wood are reduced if there are no large trees left in the stand to contribute to the CWD component (Spies et al. 1988). Downed wood is capable of retaining water, buffering temperature change, and it can provide a source of invertebrates for food. Several authors have observed lower species richness and/or lower abundance of salamanders in clearcut and young stands compared to older 35 stands (Blymer and McGinnes 1977; Pough et al. 1987; Ash 1988; Petranka et al. 1994; Dupuis etal. 1995). These patterns have been associated with reduced amounts of coarse woody debris in managed stands (Petranka et al. 1994), differences in temperature and moisture regimes between stands (Blymer and McGinnes 1977), and the loss of moist, cool microhabitats. Because of variation in the physiology and ecology of different amphibian species, variable responses to habitat changes following logging are expected. Bury et al. (1991) found that species assemblages of amphibians in unmanaged Douglas-fir forests differed across a range of stand ages and moisture regimes in the Pacific Northwest of the United States. Some species were associated with a particular stand age, while others were influenced more by moisture regime or physiographic factors. This variation in response makes it difficult to extrapolate the results of studies from one ecosystem to another. Long-toed salamanders (Ambystoma macrodactylum) occupy a variety of terrestrial habitat types, including several forest types, semi-arid areas and alpine meadows (Ferguson 1961; Orchard and Harcombe 1988). They are found from sea-level up to 3000 m in elevation. Adults are almost completely terrestrial outside of the breeding season, residing at various distances from water (Orchard 1984). Long-toed salamanders are often difficult to find outside of the breeding season (Ferguson 1961), and may lead a primarily subterranean existence during unfavourable weather. The general habitat requirements of long-toed salamanders in terrestrial habitats include cover from desiccation and temperature extremes, foraging opportunities, and overwintering sites (Orchard 36 1988). Cover is commonly found under forest floor litter, in the abandoned burrows of small mammals, and under downed wood (Green and Campbell 1984; Orchard 1984, 1988). The degree to which long-toed salamanders use downed wood compared to burrows and other cover types is not known. Long-toed salamanders are a pond-breeding species with a wide distribution in BC. They are considered habitat generalists, but have a physiology common to other amphibians and may therefore be affected by removal of forest canopy in terrestrial habitats. This study was designed to test the hypothesis that the relative abundance of long-toed salamanders is reduced in patch cuts compared to uncut forest, and that the retention of downed wood on the ground following harvesting may reduce this effect by providing long-toed salamanders with cover from otherwise unsuitable conditions. Methods Experimental Design I measured the relative abundance of salamanders in 1.7-ha experimental units to which combinations of harvesting and downed wood treatments were applied in a split-plot design. The main plot factor was harvesting treatment, with two levels: complete removal of the forest canopy in 1.7-ha patch cuts (three over a 20-ha area with additional smaller patch cuts as well) and no canopy removal over the entire 20-ha area. The split-plot factor was volume of downed wood ('low' or 'high'), applied across a patch cut or to an equivalent area within the uncut forest. This design was replicated at two sites, Mud Lake and Opax. 37 Manipulations of downed wood levels were conducted under the direction of Vanessa Craig (graduate student, Faculty of Forestry, University of British Columbia), for her study of the ecology of small mammals and downed wood. Downed wood treatments were randomly assigned to the three treatment units within each harvesting treatment. In patch cut treatment units, all wood > 6 cm in diameter was removed from the 'low' treatment units. Logs with a decay class of 4 or 5 were broken apart and scattered because they could not be moved. Extra downed wood from the surrounding area was added to the 'high' units. In forested treatment units, 'low' downed wood units were manipulated in a manner similar to that for patch cuts. 'High' downed wood in forested units was equivalent to natural densities of downed wood. Amphibians were sampled in grids of pitfall traps at one to two locations within each treatment unit. The center of each grid was 40 m from two edges of the unit. Locations were randomly selected from the four possibilities (NW, NE, S E , SW), except that a corner was occasionally eliminated when it was impossible to install traps (due to high water table or rocky substrate). Trapping Within each experimental unit, long-toed salamanders were captured using arrays of drift fences with pitfall traps in 1995, and in 1996 using grids of pitfall traps without drift fences. Because drift fences could interfere with the movement patterns of small mammals being studied in these areas, it was necessary to lift fences between trapping sessions. This was very labour 38 intensive, outweighing the benefit of using fences, so I used grids of pitfall traps without drift fences in 1996. The 1995 arrays were made from 19.6 m of 30-cm high plastic fencing (10 cm below ground and 20 cm above) in an "X" design, with four pitfall traps in the centre and eight traps at the ends of its arms. One of my aims in selecting a pitfall trap design was to minimize captures of small mammals, both to prevent small mammal mortality and to eliminate the threat they may pose to amphibians. I was able to avoid this problem by using a shallower trap than that typically recommended for capturing amphibians. The pitfall traps used in the 1995 arrays were made from plastic beer cups. Low numbers of adult captures lead me to suspect escape was possible for larger salamanders. In 1996, I moved to a deeper trap (12.5 cm) with a more effective funnel lid. The 750-mL plastic container was sufficient to capture fully grown salamanders when accompanied by this funnel lid. Captures of small mammals were limited to very small individuals, unless the trap had retained water (in which case there was occasional small mammal mortality due to drowning). Wet moss or woody debris was placed in the trap bottom to protect salamanders from desiccation and predators. Pitfall traps were installed in grids of 25 traps (5 x 5), spaced at intervals of 4 m. There were two four-day trapping sessions in 1995, one in September and one in October. In 1996, there were two summer trapping sessions (late June, early August) and two back-to-back trapping sessions in the fall (early October and mid-October). Trapping sessions ranged in length from four to eight days in 39 1996, consistent within sessions. Two grids in each unit were sampled in the late June and early August sessions. One of the two grids per unit was randomly selected for sampling in the October sessions. For each salamander captured, I recorded the snout-to-vent length (SVL) and total length to the nearest centimetre. SVL was measured from the tip of the snout to the anterior end of the vent opening (approximated as the point where the posterior of the back legs connect). Weight, to the nearest 0.1 g, was measured using a Pesola scale. Sex was recorded for adult salamanders, and any injuries, such as tail damage and deformations, were noted for each animal captured. Sketches of the dorsal pattern of long-toed salamanders were used to identify recaptured individuals. Salamanders were released near their capture location. Habitat Sampling I measured various habitat attributes on pitfall grids, some for the overall grid and others by individual trap location (Table 5). I estimated the percent cover of various attributes around each pitfall trap using 0.5-m radius circular plots. I measured the nearest piece of downed wood to each trap (within 2 m), and counted the number of pieces within 1 m of each trap. Downed wood volume for the grid was measured along a 45-m transect (four sides of a square with corners at the middle outside traps in the grid). For each piece >= 7.5 cm in diameter, I recorded species, diameter, decay class, and height above ground. I measured the distance between the center of each grid and the nearest 40 Table 5. Habitat attributes measured at terrestrial pitfall grids. Attribute Description Recorded for overall grid: Downed wood volume3 Downed wood number3 Slope Aspect Distance to water Recorded at individual traps: Canopy cover Nearest downed wood3 Nearest tree recorded along a 45-m transect recorded along a 45-m transect in degrees in degrees distance from center of grid to nearest potential breeding pond moosehorn device at eye-level over trap distance (cm), species, decay class, diameter (cm) distance (cm), species, profile class, dbh (cm) Recorded in a 0.5-m radius circular plot around each trap: Shrub cover Herb cover Grass cover Ground cover Downed wood a Sticks b Decayed wood 0 Moss Deciduous leaf litter Needles Organic materiald Exposed mineral soil Rock Slope % cover and dominant species % cover and dominant species % cover and dominant species % cover of each ground cover attribute in degrees 3 pieces with diameter 7.5 cm and greater b pieces with diameter < 7.5 cm c loose decayed woody material d undisturbed forest floor 41 potential breeding pond (in most cases on the ground; some by map estimation). Grid-level attributes were measured at all grids. Trap-level attributes were measured at a sub-sample of grids at Opax. Analyses Calculation of capture rates Capture rates for 1995 are expressed as number of salamanders per four array-nights (equivalent to one trapping session). For 1996, capture rates are expressed as number of salamanders per 100 trap-nights. This allows for comparison of rates between sessions, and makes the magnitude of the rate most similar to the actual number of animals captured at one grid over four days. I calculated mean values for each unit from the two sessions in each season. Analysis of treatment effects I tested for differences and interactions between sites, and harvesting and downed wood treatments using ANOVA, with seasons analyzed separately. Analysis for the summer sessions is based on two grids per experimental unit (used as repeated observations); for fall it is based on one grid per unit. Long-toed salamanders were assigned to two age groups (juvenile, adult), based on visual characteristics and, in cases where the age group was unclear, the discriminant function developed from salamanders captured at pond-side drift fences (see Chapter 3). Analyses were run again on the separate age groups. I tested for normality using normal probability plots, and homogeneity of variance 42 using the F-max test. Variables were transformed when necessary to meet assumptions. I used paired t-tests to compare summer rates and fall rates within forested and patch cut units, separately for each age group. Habitat Associations Correlations between residual capture rates (minus mean rate for site) and grid-level habitat attributes (slope, volume of downed wood, number of pieces of downed wood, minimum distance to breeding ponds) were run by season and age group. T-tests were used to compare trap-level habitat attributes between forest and patch cut treatments, using 12 Mud Lake grids and 6 Opax grids for each test. Percent covers of exposed rock and exposed mineral soil were combined into one variable for the analysis. The Bonferroni adjustment was applied to adjust p-values for the use of multiple t-tests. Results Effects of Treatments on Capture Rates of Long-toed Salamanders a) Summer Trapping Sessions- 1996 There was a significant interaction (F (i i8) = 9.8; p = 0.01) between site and harvesting treatment for the summer session. Comparisons of treatment means indicated that the mean capture rate in forested units at Mud Lake was significantly higher than that in the patch cut units, and in both treatment types at the Opax site (Figure 8). Results for site and harvesting treatment (Table 6) are difficult to interpret due to this interaction. Analysis by age group did not indicate 43 Forest Patch cut Harvesting treatment Figure 8. Mean capture rates of long-toed salamanders (LTS) for the 1996 summer trapping sessions in forest and patch cut treatment units by site. Error bars are 1 S E . 44 Table 6. Results of ANOVA comparing harvesting and downed wood treatments for the 1996 summer trapping sessions. Source of Sum of Mean Variation df Squares Squares F p main plot site 1 3.240 3.240 26.182 0.001 harvesting 1 1.823 1.823 1.507 > 0.10 site x harvesting 1 1.210 1.210 9.778 0.01 split plot clowned wood 1 0.022 0.022 0.138 > 0.10 harvesting x downed wood 1 0.023 0.023 2.30 > 0.10 site x downed wood 1 0.160 0.160 1.293 > 0.10 site x harvesting x downed wood 1 0.010 0.010 0.081 > 0.10 error 8 0.990 0.124 45 any significant treatment differences (Table 7), but the trend of higher capture rates in forest versus patch cut units at Mud Lake was consistent for both adult and juvenile salamanders (Figure 9). There was a single capture at the Opax grids in the summer sessions, one adult salamander in the high downed wood-uncut forest treatment unit. There were no differences between downed wood treatments in the summer season (Table 6); again the single capture at Opax limited my ability to detect differences. In forested units at Mud Lake, capture rates were similar in high and low grids for both age groups (Figure 10). In the patch cut units, both of the two adults captured were in the low units and the single juvenile capture was in the high downed wood unit. b) Fall Trapping Sessions-1996 During the fall trapping sessions of 1996, mean capture rates for adult long-toed salamanders were similar between patch cut and forested units (0.65 and 0.68 LTS /100 trap-nights, respectively) ( F ( 1 1 ) = 0.027; p = 0.90) (Figure 11a). Juvenile capture rates in patch cut units appeared greater than those in forested units (1.45 vs. 0.56 LTS /100 trap-nights), but this difference was not statistically significant (F(i,i) = 15.91; p = 0.16) (Figure 11b). I found no differences between high and low downed wood treatments for adult salamanders ( F ( i i ) = 47.25; p = 0.10; Figure 12a). Adult capture rates appeared greater in the low downed wood units within the patch cut treatment blocks compared to high, but were similar between downed wood treatments 46 Table 7. Results of ANOVAs comparing harvesting and downed wood treatments for the 1996 summer trapping sessions, by age group. a) Adult long-toed salamanders Source of Sum of Mean Variation df Squares Squares F p main plot site 1 1.44 1.44 12.66 0.007 harvesting 1 0.90 0.90 1.84 > 0.10 site x harvesting 1 0.49 0.49 4.31 0.07 split plot downed wood 1 0.02 0.02 0.14 >0.10 harvesting x downed wood 1 0.12 0.12 12.30 > 0.10 site x downed wood 1 0.16 0.16 1.41 >0.10 site x harvesting x downed wood 1 0.01 0.01 0.09 > 0.10 error 8 0.91 0.11 b) Juvenile long-toed salamanders Source of Sum of Mean Variation df Squares Squares F p main plot site 1 0.36 0.36 3.60 0.09 harvesting 1 0.16 0.16 1.00 > 0.10 site x harvesting 1 0.16 0.16 1.60 > 0.10 split plot downed wood 1 0.00 0.00 1.00 >0.10 harvesting x downed wood 1 0.04 0.04 1.00 >0.10 site x downed wood 1 0.00 0.00 1.00 >0.10 site x harvesting x downed wood 1 0.04 0.04 0.40 > 0.10 error 8 0.80 0.10 47 Forest Patch cut Harvesting treatment Figure 9. Mean capture rates of long-toed salamanders (LTS) for the 1996 summer sessions in forest and patch cut treatment units at the Mud Lake site, by age group. Error bars are 1 S E . 48 a) Adults High Low High Low Downed wood treatment b) Juveniles High Low High Low Downed wood treatment Figure 10. Mean capture rates of long-toed salamanders (LTS) for the 1996 summer sessions in high and low downed wood treatment units at Mud Lake by harvesting unit, for a) adult salamanders and b) juvenile salamanders. Error bars are 1 S E . 49 a) Adults Mud Lake Opax x: O) 'E 6. 1.5 1 Forest Patch cut Forest Patch cut Harvesting treatment b) Juveniles 3.5 £ 3 cn A. " 2 2 o o 1.5 |2 0.5 Mud Lake Opax Forest Patch cut Forest Harvesting treatment Patch cut Figure 11. Mean capture rates of long-toed salamanders (LTS) for the fall trapping sessions during 1996 in forest and patch cut treatment units by site, for a) adult salamanders and b) juvenile salamanders. Error bars are 1 S E . 50 a) Adults Forest Patch cut I •j? 1.5 • a. n 0 °- 0.5 to I High Low High Downed wood treatment Low b) Juveniles 3.5 | 3 co c 2.5 • a. g 2 o ° 1.5 1 0.5 0 CD a. tn Forest Patch cut -+-High Low High Downed wood treatment Low Figure 12. Mean capture rates of long-toed salamanders (LTS) for fall sessions during 1996 in high and low downed wood treatment units by harvesting unit, for a) adult salamanders and b) juvenile salamanders. Error bars are 1 S E . 51 within forested units (Figure 12a), although this apparent interaction was not statistically significant (F (ii) = 2.261; p = 0.37). No difference in capture rates between downed wood treatments was found for juvenile salamanders (F (ii) = 0.427; p > 0.10; Figure 12b). c) Fall Trapping Sessions-1995 Most salamanders captured in the fall of 1995 were juveniles, possibly due in part to problems with the trap design. The eight adult salamanders captured were found in equal numbers at each site, and in a variety of treatment units. Analysis of juvenile capture rates did not indicate significant differences between harvesting treatments ( F ( 1 1 ) = 0.166; p = 0.80), or downed wood treatments (F(2,2) = 0.670; p > 0.20), but there was only one capture at the Opax site. Juvenile capture rates at the Mud Lake site were highest in the low downed wood treatment units for both the forest and patch cut treatments, and rates were somewhat higher in the patch cut treatments overall (Figure 13). d) Between Seasons and Sites In 1996, mean capture rates in the fall sessions were 3.3 times that of summer sessions (mean of all grids and sites). There was a larger difference between the mean capture rates of summer and fall trapping sessions in patch cut units than in forested units, although none of the these differences were statistically significant (Figure 14). For patch cut units, the mean capture rate for 52 Figure 13. Mean capture rates of juvenile long-toed salamanders (LTS) at downed wood arrays at Mud Lake in 1995, by harvesting treatment. a) Adults Summer Fall b) Juveniles Figure 14. Change in mean capture rates between summer and fall trapping sessions during 1996 by harvesting treatment, for a) adult long-toed salamanders (LTS) and b) juvenile salamanders. Error bars are 1 S E . 54 adult long-toed salamanders in summer sessions was 0.13 LTS/100 trap-nights, compared to 0.65 LTS/100 trap-nights in the fall (t = 2.026, df = 3, p = 0.14). Adult capture rates in each season were more similar in forested units (0.60 and 0.68 LTS/100 trap-nights for summer and fall, respectively) (t = 0.629, df = 3, p = 0.57). For juvenile salamanders in patch cut units, the mean capture rate was 0.05 LTS/100 trap-nights in the summer and 1.45 LTS/100 trap-nights in the fall (t = 2.217, df = 3, p = 0.11). Rates for juveniles in forested units were 0.25 LTS/100 trap-nights in the summer, and 0.56 LTS/100 trap-nights in the fall (t = 0.686, df = 3, p = 0.54). Capture rates were consistently higher at the Mud Lake site than at the Opax site, but this difference was only statistically significant in the fall trapping sessions of 1996 (t = 3.483, df = 6, p = 0.01). The trend of higher capture rates at the Mud Lake site was observed for both adult and juvenile salamanders (Figure 15). Effects of Treatments on Individual Salamanders Overall capture rates in the summer were too small to allow comparisons of condition between harvesting treatments during that season. Rates at the Opax site were also too low to allow comparison. I did not pool site data because of differences in mean SVL between sites. For the comparison of condition between treatments for the fall trapping sessions, I excluded six individuals with less than certain age grouping (five juveniles and one adult 55 a) Adults 1996 Summer Fall 1.5 J2 x: O) c D. 1 + 0.5 5 Mud Lake Opax Mud Lake Opax Site b) Juveniles 1996 Summer Fall 2.5 J3 S 2 c i . 5 1.5 « 0.5 Mud Lake Opax Mud Lake Opax Site c) Juveniles 1995 3 T I 2.5 CO ? 2 >> 2 (5 1.5 I 1 ^ 0.5 -+-Mud Lake Opax Site Figure 15. Comparison of mean capture rates at the Mud Lake and Opax sites by season, for a) adult long-toed salamanders (LTS) in 1996, b) juvenile salamanders in 1996, and c) juvenile salamanders in 1995. Error bars are 1 S E . 56 based on the discriminant function classification). Adult salamanders captured in the fall in forested units at the Mud Lake site had a higher mean weightSVL ratio than those in patch cut units (t = 2.22, df = 14, p = 0.04). There was no difference in weightSVL ratio for juveniles (t = 1.02, df = 19, p = 0.32). Habitat In addition to the obvious differences in canopy cover, the percent cover of six vegetation and ground cover habitat attributes differed significantly between forest and patch cut treatments (Table 8). The percent cover of grass, sticks and rock and exposed mineral soil was higher in patch cut treatments compared to forested treatments (grass, t = -5.48, df = 16, p = 0.001; sticks, t = -3.19, df = 16, p = 0.06; rock and mineral soil, t = -4.83, df = 16, p = 0.002). Values for moss, deciduous leaf litter and needles were higher in forested than patch cut treatments (moss, t = 4.58, df = 16, p = 0.003; leaf litter, t = 7.23, df = 16, p < 0.001; needles, t = 3.36, df = 16, p = 0.04). Although there was some variation in the percent cover of these attributes between sites, in most cases differences were not significant. Needle cover in forested units at Opax was significantly higher than in forested units at Mud Lake (t = -11.07, df = 7, p < 0.001). Percent cover of organic material also tended to be higher at Opax in forested units, but this difference was not significant after application of the Bonferroni adjustment for multiple comparisons (t = -3.06, df = 7, p = 0.22). Shrub cover in patch cut units tended to be higher at Mud Lake than at Opax, but 57 Table 8. Percent cover of various habitat attributes by site and block: mean (SE); Mud Lake, n = 6 grids; Opax Mountain, n = 6 grids for canopy and n = 3 for all other attributes. Attribute Mud Lake Opax Mountain Forest Patch Cut Forest Patch Cut Canopy cover 16.5 (0.9) 0.0 (0.0) 14.4 (1.5) 0.0 (0.0) Shrub cover 14.3 (2.2) 17.4 (2.3) 19.8 (3.3) 7.2 (0.7) Herb cover 26.7 (2.9) 35.4 (3.2) 20.3 (2.1) 24.9 (6.3) Grass cover 14.5 (3.5) 47.7 (3.0) 24.8 (2.1) 36.8 (9.4) Downed wood 7.8(1.7) 5.4(1.9) 6.0 (3.5) 13.9 (7.1) Sticks 14.7(1.7) 25.1 (4.4) 18.3 (3.1) 41.7(6.2) Decayed wood 4.5(2.2) 0.7 (0.4) 0.7 (0.7) 0.0 (0.0) Moss 27.2 (5.5) 4.0 (1.5) 16.2 (5.6) 1.8 (0.6) Deciduous leaf litter 22.1 (3.2) 1.7 (0.3) 36.9 (5.3) 0.0 (0.0) Needles 7.1 (1.5) 0.3 (0.2) 33.8(1.6) 1.0 (0.7) Organic material 12.0 (3.5) 13.3 (1.2) 28.6 (2.7) 15.6 (2.5) Rock and exposed soil 1.6 (0.6) 8.3 (1.3) 0.1 (0.1) 12.4 (4.6) 58 again this difference was not significant after application of the Bonferroni adjustment (t = 3.07, df = 7, p = 0.22). No significant correlations were detected between capture rates and grid-level habitat attributes, for either age group or season. Minimum distances between grids and potential breeding ponds ranged from 47-160 m. Discussion The results of this study do not support the hypothesis that long-toed salamander abundance is reduced in patch cut areas compared to uncut forests. Very low capture rates at one of the two replicate sites, limited my ability to detect any potential treatment effects. Patterns of relative abundance between treatments at the Mud Lake site suggest that salamander abundance may be reduced in patch cut areas during the summer compared to forested areas. Weather conditions later in the fall appear to be more suitable for the activity of long-toed salamanders in patch cut areas. Removal of the forest canopy results in temperature extremes (and/or wider fluctuations), and can reduce humidity and soil moisture (Chen et al. 1993). Soil temperature measurements by other researchers on this site showed elevated temperatures in patch cuts compared to the adjacent forest, at depths of 1 and 15 cm, with a greater contrast (4-6°C) between the patch cut and adjacent forest observed closer to the surface (Miege et al. 1998). Soil temperatures at a depth of 15 cm in the patch cuts were comparable to those at the surface (1 cm) in the adjacent forest. Differences in ground cover between 59 treatments included greater cover of moss and deciduous leaf litter in forested units compared to patch cuts. Long-toed salamanders were observed using both of these attributes as cover. Juvenile salamanders are more at risk of desiccation than adults due to their smaller size. The only summer captures of juvenile salamanders were in June, predominantly in the forest. Patterns of juvenile abundance among harvesting treatments were similar in the fall for 1995 and 1996. There was a tendency toward higher rates for juveniles in patch cuts compared to forested units. When temperatures are cooler in the fall, juvenile salamanders may require more open habitats to keep their temperature in the appropriate range. Pough (1983) cites an example of three species of Ameiva lizards where body size was a factor in habitat use, due to differences in their rates of cooling. The smallest species foraged in more open habitats than the larger ones, and juveniles of the larger species behaved like adults of smaller species, selecting more open habitats for foraging. There was no indication that downed wood was important in patch cuts to mediate any negative impacts of canopy removal. However, this should not be interpreted as a lack of importance of downed wood to long-toed salamanders. Salamanders were observed under or in downed wood on several occasions over the course of the study. One reason not to draw conclusions about the importance of downed wood follows from the problems with low sample size. A second reason relates to application of the 'low' downed wood treatments. Large, well-decayed logs were difficult to remove, and so were pulled apart on 60 the site. While this removed the structure of the downed wood, it resulted in areas of loose, well-decayed woody debris which could potentially provide excellent habitat for salamanders. Long-toed salamanders were found in this material on a couple of occasions. 61 Chapter 4. Conclusions and Recommendations The small wetlands sampled in this study support good-sized populations of long-toed salamanders, and represent an important habitat feature for this species and other pond-breeding amphibians. Both permanent and ephemeral ponds were important for breeding, although ponds that retained water longer were used more often than those ponds that dried very early in the summer. Ephemeral ponds may actually be more suitable for breeding than larger, permanent wetlands and lakes, because of the warmer, shallow habitats they often provide and the absence of certain predators. Results of this study suggest that removal of the canopy cover adjacent to breeding sites may reduce breeding populations of long-toed salamanders over time, although ponds with low canopy cover are still used for breeding. In addition to reducing the number of available breeding sites, the loss of small wetlands as suitable breeding habitat would lead to increased inter-wetland distances, influencing the potential for recovery of populations by limiting immigration from other sites. Gibbs (2000) recommends that any size-based program for wetland protection should have a lower limit of no more than 0.4 ha to adequately maintain wetland biota. At size limits of 0.8 ha and greater, inter-wetland distances would be greater than the maximum distances travelled by most amphibians. Riparian management guidelines in BC use a 1 ha minimum for most biogeoclimatic zones, and 0.5 and 0.25 ha minimum for some of the drier subzones where wetlands are less abundant. The ponds sampled in this study were all less than 0.25 ha in size, too small to require the minimum level of 62 protection under current guidelines for riparian management, yet they area able to support substantial populations of long-toed salamanders. Some consideration of these smaller wetlands needs to be incorporated into forest management. This may be especially important in areas with few wetlands, where there is little opportunity for recolonization following population reductions. Information on the time it takes for habitats to recover is needed to assess the level of protection that small wetlands require. There was some indication that canopy removal in terrestrial habitats may be restricting the abundance and/or activity of long-toed salamanders in patch cut areas, at least during summer months. Results of the terrestrial component of the study were not conclusive, due to very low sample sizes at one site. The design of this study could have been improved by first examining salamander abundance in relation ecosystem type, and applying treatments within habitat types that support populations of a reasonable size. A better understanding of how the relative abundance of long-toed salamanders varies between ecosystem types would also improve our ability to manage for this species. Improved knowledge of the particular microhabitats used by long-toed salamanders could lead to suggestions for mitigation of the effects of harvesting, through efforts to minimize impacts on important habitat features. For the protection of pond-breeding species like the long-toed salamander, management of wetland habitats should be considered together with management of the adjacent terrestrial habitat. Semlitsch (1998) estimated that buffer zones extending 165 m from the wetland edge into the terrestrial 63 habitat would account for 95% of the population around a given breeding site, for several salamanders of the genus Ambystoma. This strategy may be appropriate for areas with a few large wetlands, but would be less practical as a means of protecting smaller aggregated wetlands. Arrays of small patch cuts may offer a practical means of protecting small wetlands, as their locations can be considered in the layout of patch cut openings. The use of small opening sizes like those at the Opax Mountain research site (0.1 to 1.6 ha), also provides uncut areas in close proximity, allowing long-toed salamanders to move out of openings into uncut areas during times of unfavourable weather. 64 Literature Cited Adams, M. J . , K. O. Richter, and W. P. Leonard. 1997. Surveying and monitoring amphibians using aquatic funnel traps. Pages 47-54 in D. H. Olsen, W. P. Leonard and R. B. Bury, eds. Sampling amphibians in lentic habitats. Society for Northwestern Vertebrate Biology, Olympia, Washington. Anderson, J . D. 1967. A comparison of life histories of coastal and montane populations of Ambystoma macrodactylum in California. Amer. Midi. Nat. 77:323-335. Ash, A. N. 1988. Disappearance of salamanders from clearcut plots. J . Elisha Mitchell Sci. Soc. 104:116-122. Berven, K. A. 1990. Factors affecting population fluctuations in larval and adult stages of the wood frog (Rana sylvatica). Ecology 71:1599-1608. Blaustein, A. R., and D. B. Wake. 1990. Declining amphibian populations: A global phenomenon? Trends Ecol. Evol. 5:203-304. Blaustein, A. R., P. D. Hoffman, D. G. Hokit, J . M. Kiesecker, S. C. Walls, and J. B. Hays. 1994. UV repair and resistance to solar UV-B in amphibian eggs: A link to population declines? Proc. Natl. Acad. Sci. 91:1791-1795. Blymer, M. J . , and B. S. McGinnes. 1977. Observations on possible detrimental effects of clearcutting on terrestrial amphibians. Bull. Maryland Herp. Soc. 13:79-83. British Columbia Ministry of Forests and B.C. Environment. 1995. Riparian Management Area Guidebook. Forest Practices Code of British Columbia. B.C. Ministry of Forests and B.C. Environment, Victoria, B.C. 68pp. Bury, R. B., P. S. Corn, K. B. and Aubry. 1991. Regional patterns of terrestrial amphibian communities in Oregon and Washington. Pages 341-362 in L. F. Ruggiero, K. B. Aubry, A. B. Carey, and M. H. Huff, eds. Wildlife and Vegetation of Unmanaged Douglas-fir Forests. Gen. Tech. Rep. PNW-GTR-285. U.S. Dept. Agr., For. Ser., Pacific Northwest Res. Stn., Portland, OR. Chen, J . , J. F. Franklin, T. A. and Spies. 1993. Contrasting microclimates among clearcut, edge, and interior of old-growth Douglas-fir forest. Agric. Forest Meteorol. 63:219-237. 65 Clark, K. L. 1986. Responses of spotted salamander, Ambystoma maculatum, populations in central Ontario to habitat acidity. Can. Field-Nat. 100:463-469. Corn, P. S., and R. B. Bury. 1989. Logging in western Oregon: responses of headwater habitats and stream amphibians. For. Ecol. Manage. 29:39-57. Crump, M. 1989. Effect of habitat drying on developmental time and size at metamorphosis in Hyla pseudopuma. Copeia 1989:794-797. Dale, J. M., B. Freedman, and J. Kerekes. 1985. Acidity and associated water chemistry of amphibian habitats in Nova Scotia. Can. J. Zool. 63:97-105. Davis, T. M. 1999. Study designs for evaluating the effects of forestry activities on aquatic-breeding amphibians in terrestrial forest habitats of British Columbia. Minist. Environ., Lands and Parks, Wildl. Branch, Victoria, B.C. Wildl. Working Rep. No. WR-97. 42pp. Dupuis, L. A., J . N. M. Smith, and F. Bunnell. 1995. Relation of terrestrial-breeding amphibian abundance to tree-stand age. Cons. Biol. 9:645-653. Enge, K. M., and W. R. Marion. 1986. Effects of clearcutting and site preparation on herpetofauna of a north Florida flatwoods. For. Ecol. Manage. 14:177-192. Ferguson, D. E. 1961. The geographic variation of Ambystoma macrodactylum Baird, with the description of two new subspecies. Amer. Midi. Nat. 65: 311-338. Freda, J . , and W. A. Dunson. 1986. Effects of low pH and other chemical variables on the local distribution of amphibians. Copeia 1986:454-466. Gibbs, J . P. 1993. Importance of small wetlands for the persistence of local populations of wetland-associated animals. Wetlands 13:25-31. Gibbs, J . P. 2000. Wetland loss and biodiversity conservation. Cons. Biol. 14:314-317. Green, D. M., and R. W. Campbell. 1984. The Amphibians of British Columbia Handbook No. 45. B.C. Provincial Museum, Victoria. Holomuzki, J . R. 1986. Predator avoidance and diel patterns of microhabitat use by larval tiger salamanders. Ecology 67:737-748. 66 Howard, J . H., and R. L. Wallace. 1985. Life history characteristics of populations of the long-toed salamander (Ambystoma macrodactylum) from different altitudes. Amer. Midi. Nat. 113:361-373. Huggard, D., R. Walton, and W. Klenner. 1998. Depth and duration of snow at the Opax Mountain Silvicultural Systems Site. Pages 186-193 in A. Vyse, C. Hollstedt, and D. Huggard, eds. Managing Dry Douglas-fir Forests of the Southern Interior: Workshop Proceedings. April 29-30, 1997. Kamloops, British Columbia, Canada. Res. Br., B.C. Min. For., Victoria, B.C. Work Pap. 34/1998. Klenner, W., and A. Vyse. 1998. The Opax Mountain Silvicultural Systems project: Evaluating alternative approaches to managing dry Douglas-fir forests. Pages 128-135 in A. Vyse, C. Hollstedt, and D. Huggard, eds. Managing Dry Douglas-fir Forests of the Southern Interior: Workshop Proceedings. April 29-30, 1997. Kamloops, British Columbia, Canada. Res. Br., B.C. Min. For., Victoria, B.C. Work Pap. 34/1998. Leips, J . , and J. Travis. 1994. Metamorphic responses to changing food levels in two species of hylid frogs. Ecology 75:1345-1356. Licht, L. 1974. Survival of embryos, tadpoles, and adults of the frogs Rana aurora aurora and Rana pretiosa pretiosa sympatric in southwestern British Columbia. Can. J. Zool. 52:613-627. Licht, L. 1992. The effect of food level on growth rate and frequency of metamorphosis and paedomorphosis in Ambystoma gracile. Can. J . Zool. 70:87-93. Miege, D., D. Lloyd, and A. Arsenault. 1998. Spatial and temporal response of vegetation to silvicultural practices in Interior Douglas-fir forests. Pages 211-225 in A. Vyse, C. Hollstedt, and D. Huggard, eds. Managing Dry Douglas-fir Forests of the Southern Interior: Workshop Proceedings. April 29-30, 1997. Kamloops, British Columbia, Canada. Res. Br., B.C. Min. For., Victoria, B.C. Work Pap. 34/1998. Newman, R. A. 1994. Effects of changing density and food level on metamorphosis of a desert amphibian, Scaphiopus couchii. Ecology 75:1085-1096. Orchard, S. A. 1984. Amphibians and Reptiles of B.C.: An Ecological Review. Research Branch, Ministry of Forests. WHR-15. Victoria, B.C. Orchard, S. A. 1988. Wildlife Habitat Handbooks for the Southern Interior Ecoprovince, vol. 4: Species notes for amphibians. Ministry of Environment, WHR-35. Victoria, B.C. 67 Orchard, S. A., and A. P. Harcombe. 1988. Wildlife Habitat Handbooks for the Southern Interior Ecoprovince, vol. 8: species-habitat relationship models for amphibians. Ministry of Environment, WHR-35. Victoria, B.C. Pechmann, J. H. K., D. E. Scott, J . W. Gibbons, and R. D. Semlitsch. 1989. Influence of wetland hydroperiod on diversity and abundance of metamorphosing juvenile amphibians. Wetlands Ecol. Manage. 1:3-11. Petranka, J . W. 1984. Sources of interpopulational variation in growth responses of larval salamanders. Ecology 65:1857-1865. Petranka, J . W. 1989. Density-dependent growth and survival of larval Ambystoma: Evidence from whole-pond manipulations. Ecology 70:1752-1767. Petranka, J . W., M. P. Brannon, M. E. Hopey, and C. K. Smith. 1994. Effects of timber harvesting on low elevation populations of southern Appalachian salamanders. For. Ecol. Manage. 67:135-147. Petranka, J. W., and A. Sih. 1986. Environmental instability, competition, and density-dependent growth and survivorship of a stream-dwelling salamander. Ecology 67:729-736. Pough, F. H. 1983. Amphibians and reptiles as low-energy systems. Pages 141-188 in W. P. Aspey and S. I. Lustick, eds. Behavioural Energetics: The Cost of Survival in Vertebrates. Ohio State University Press, Columbus, Ohio. Pough, F. H., E. M. Smith, D. H. Rhodes, and A. Collazo. 1987. The abundance of salamanders in forest stands with different histories of disturbance. For. Ecol. Manage. 20:1-9. Raphael, M. G. 1988. Long-term trends in abundance of amphibians, reptiles and mammals in Douglas-fir forests of Northwestern California. Pages 23-31 in R. C. Szaro, K. E. Severson, and D. R. Patton, eds. Management of Amphibians, Reptiles and Small Mammals in North America. Proceedings of the Symposium, Flagstaff, Arizona, July 1988. Richter, K. O. and A. L. Azous. 1995. Amphibian occurrence and wetland characteristics in the Puget Sound Basin. Wetlands 15:305-312. Russell, A. P., G. L. Powell, and D. R. Hall. 1996. Growth and age of Alberta long-toed salamanders (Ambystoma macrodactylum krausei): a comparison of two methods of estimation. Can. J. Zool. 74: 397-412. 68 Scott, D. E. 1994. The effect of larval density on adult demographic traits in Ambystoma opacum. Ecology 75:1383-1396. Semlitsch, R. D., and H. M. Wilbur. 1988. Effects of pond drying time on metamorphosis and survival in the salamander Ambystoma talpoideum. Copeia 1988:978-983. Semlitsch, R. D. 1998. Biological delineation of terrestrial buffer zones for pond-breeding salamanders. Cons. Biol. 12:1113-1119. Shoop, C. R. 1974. Yearly variation in larval survival ot Ambystoma maculatum. Ecology 55:440-444. Skelly, D. K. 1995. A behavioural trade-off and its consequences for the distribution of Pseudacris treefrog larvae. Ecology 76:150-164. Spies, T. A., J . F. Franklin, and T. B. Thomas. 1988. Coarse woody debris in Douglas-fir forests of western Oregon and Washington. Ecology 69:1689-1702. Walls, S. C , A. R. Blaustein, and J. J . Beatty. 1992. Amphibian biodiversity of the Pacific Northwest with special reference to old-growth stands. Northwest Env. J . 8:53-69. Welsh, H. H., and A. J. Lind. 1988. Old growth forests and the distribution of terrestrial herpetofauna. Pages 439-455 in R. C. Szaro, K. E. Severson, and D. R. Patton, eds. Management of Amphibians, Reptiles and Small Mammals in North America. Proceedings of the Symposium, Flagstaff, Arizona, July 1988. Wilbur, H. M. 1980. Complex life cycles. Ann. Rev. Ecol. Syst. 11:67-93. 69 

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:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0090036/manifest

Comment

Related Items