Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Patterns of nest predation and nest predator abundance in a fragmented englemann spruce/subalpine fir… Campbell, Victoria Elizabeth Anne 1995

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

Item Metadata

Download

Media
831-ubc_1995-0325.pdf [ 2.24MB ]
Metadata
JSON: 831-1.0086821.json
JSON-LD: 831-1.0086821-ld.json
RDF/XML (Pretty): 831-1.0086821-rdf.xml
RDF/JSON: 831-1.0086821-rdf.json
Turtle: 831-1.0086821-turtle.txt
N-Triples: 831-1.0086821-rdf-ntriples.txt
Original Record: 831-1.0086821-source.json
Full Text
831-1.0086821-fulltext.txt
Citation
831-1.0086821.ris

Full Text

PATTERNS OF NEST PREDATION AND NEST PREDATOR ABUNDANCE IN A FRAGMENTED ENGLEMANN SPRUCE/SUB ALPINE FIR FOREST by VICTORIA ELIZABETH ANNE CAMPBELL B.Sc. (Hons.), Trent University, Peterborough, 1992 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Zoology) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA July 1995 ©Victoria Elizabeth Anne Campbell, 1995 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree i that permission for extensive copying of this thesis "for scholarly purposes may be department or by his or her representatives. It is publication of this thesis for financial gain shall not be allowed without my written permission. granted by the head of my understood that copying or Department of ^ ^ 3 3 ° L The University of British Columbia Vancouver, Canada Date \ DE-6 (2/88) A B S T R A C T 11 In a fragmented Englemann Spruce/Subalpine Fir (ESSF) forest, I used artificial nests to test the hypothesis that nest predation is greater at the forest edge and in clearcuts, than in the forest interior. In this forest type, nests in clearcuts had a lower frequency of predation than nests in the forest, yet there was no consistent difference in the frequency of predation at the forest edge or in the forest interior, a pattern opposite to that previously documented. However, in previous studies the landscape was fragmented by agricultural development, where populations of potential nest predators may be elevated as a result of anthropogenic food sources. I found no evidence that the decrease in the frequency of predation on nests in clearcuts was the result of a concomitant shift in the identity (from plasticine eggs) and/or number of predators (from predator surveys) in these locations. Rather, in habitats fragmented by logging activities, many nest predators may avoid foraging in the clearcut because of the increased predation risk to themselves in these locations. In addition to the effect of nest location on predation rate, there was marked temporal variation in the frequency of nest predation. I documented an increase in nest predation both within and between trials; and this may have been a result of predators learning to exploit artificial nests. Studies which can identify individual predators, and record their foraging activity in response to artificial nests, may shed light on the mechanisms behind this temporal variation in nest predation. While artificial nests provide an index of predation risk, future use of this method must be cautioned against until the biases can be clearly identified. Further study of the patterns of nest predation in fragmented habitats should instead focus on the patterns of nest predation on natural nests, the distribution and abundance of potential nest predators in fragmented and undisturbed forest habitats, and on how nest predators forage within clearcut, forest edge, and in forest interior habitats. T A B L E OF C O N T E N T S A B S T R A C T i i T A B L E OF C O N T E N T S i i i LIST OF TABLES v LIST OF FIGURES v i A C K N O W L E D G M E N T S v i i INTRODUCTION 1 M E T H O D S 3 Site Description 3 Patterns of Nest Predation in Forest Edge, Forest Interior, and Clearcut Habitats: Field Procedure (1993) . 4 Field Procedure (1994) 5 The Distribution and Identity of Nest Predators in the ESSF Forest: Predator Identity (1993,1994) 6 Statistical Analysis 7 RESULTS • 8 Patterns of Nest Predation in Forest Edge, Forest Interior, and Clearcut Habitats: The effect of 'Plot' and 'Time' 8 The effect of nest location... 11 The effect of nest height and cover 13 The effect of nest contents 14 The Distribution and Identity of Nest Predators in the ESSF Forest: Overall Patterns 14 Temporal and Spatial Patterns 15 iv Nest Contents 17 Patterns of Predator Distribution Across Habitats from Audi tory and Visual Surveys (1994 only) 18 Differences in predation rate between previously 'exposed' and 'unexposed' plots: A post-hoc test 19 D I S C U S S I O N 20 Patterns of Nest Predation The Effect of Fragmentation 20 The Effect of Nest Height and Cover 24 Temporal Effects 25 Artificial nests and plasticine eggs: H o w good is the evidence? 26 Management implications and future consideration 28 C O N C L U S I O N S 28 R E F E R E N C E S 37 A P P E N D I X 1 41 A P P E N D I X 2 42 LIST OF TABLES Table 1: Vegetation layers in the ESSF forest 31 Table 2: Summary of experimental design 32 Table 3: Calculation of F-statistics for ANOVA ' s 33 Table 4: A N O V A , Trial 1, 1993: frequency of nest predation 33 Table 5: A N O V A , Trial 2 and 3,1993: frequency of nest predation 34 Table 6: A N O V A , Trial 1. 1994: frequency of nest predation 34 Table 7: A N O V A (clearcuts included), Trials 1 and 2,1994: frequency of nest predation 3Z Table 8: A N O V A , Trial 1, 1993: frequency of avian and mammalian nest predators 31 Table 9: A N O V A , Trials 2 and 3, 1993: frequency of avian and mammalian nest predators 3! Table 10: A N O V A , Trial 1, 1994: frequency of avian and mammalian nest predators 3( VI LIST OF FIGURES Figure 1: Map of study area ....30 Figure 2: Effect of 'Plot' on frequency of nest predation, Trial 1, 1993 9 Figure 3: 'Plot'*'Trial' interaction on frequency of nest predation, Trials 2 and 3,1993 9 Figure 4: Average predation rate for plots, Trials 1 and 2, 1994 11 Figure 5: Effect of 'Location' on frequency of nest predation, Trials 1 and 2, 1994 12 Figure 6: 'Location'*'Plot' interaction on frequency of nest predation 13 Figure 7: Average number of avian and mammalian predators 15 Figure 8: Frequency of nests lost to mammalian predators, Trials 2 and 3, 1993 16 Figure 9: Frequency of ground and shrub nests lost to avian and mammalian predators 17 Figure 10: Average number of predators detected by auditory and visual.... surveys 19 Figure 11: Graphical model of relative frequency of predation at edge, interior and clearcut nest locations 21 vu A C K N O W L E D G M E N T S I would like to thank my supervisor, Dr. J.N.M. Smith, for his continued support throughout this project; particularly for his insight and encouragement in the final (and long) stages of the thesis. The field-work portion of this project could not have been completed without the help of my field assistants. Many thanks to: Debbie Wellwood, Patrick Doyle, Audrey Roburn, Susan Peters, and Louise Blomer. Other logistic support in the field was provided by the staff at the Weyerhaeuser office, Okanagan Falls, BC; as well as by Sim's Grocery Store (for use of their refrigerator to store my many quail eggs). Many of the statistical tests in this thesis were conducted with the help of Dr. S. Heard and Dr. W. Hochachka. Thank-you for your continued patience and advice. Special thanks to my family for their moral and (occasionally) material, support. Finally, I would like to acknowledge the financial support of the National Sciences Research Council (V.C. and J.N.M.S.), the J.K. Cooper Foundation, the Anne Vallee Memorial Fund, and Sigma Xi. INTRODUCTION 1 Habitat fragmentation has traditionally been defined as a process by which large areas of contiguous habitat, e.g. forest and prairie grassland, are reduced to isolated patches within a matrix of developed land, e.g. Wilcove et al. 1986. The patches which remain are characterized by a decreased amount of undisturbed 'core' habitat, and an increased amount of 'edge' habitat that is vulnerable to impacts from the surrounding landscape. One particular concern in forested habitat is that nest predators and brood parasites associated with the surrounding agricultural landscapes, e.g. raccoons, corvids and cowbirds, can exploit birds nesting within this edge habitat (Whitcomb et al. 1981, Brittingham and Temple 1983, Andren and Angelstam 1988). Open-nesting passerines, e.g. the wood thrush (Hylocichla mustelina), and the ovenbird (Seiurus aurocapillus), which have historically nested within unexploited forests, must cope with this increased predation pressure throughout the remaining forest habitat. This has been implicated as a major factor in their decline within North America (Robinson 1991, Robinson et al. 1995). Most past studies of forest fragmentation have been conducted in industrial and agricultural areas (Wilcove 1985, Andren et al. 1985, Robinson et al. 1995). In these areas, populations of potential nest predators and parasites may be elevated by anthropogenic food sources within the landscape, e.g. crops and refuse (Wilcove et al. 1986, Askins 1995). Thus the increases in nest predation within edge habitat may be the result of increases in the number of predators associated with the surrounding landscape, rather than by the creation of edge habitat itself. However, much of the recent forest fragmentation in North America involves logging activities well away from human settlements and agricultural developments. To what extent does the extensive network of clearcuts and roads generated by logging activities produce an increase in nest predation at forest edges? To date, only a few studies have successfully investigated the patterns of nest 2 predation in logged habitats (Small and Hunter 1988, Rudnicky and Hunter 1993). Poor success has been due to a lack of replication, and/or the confounding effects of other, unrelated treatments (Ratti and Reese 1988, Yahner et al. 1993; but see Paton (1994) and Murcia (1995) for a full review). In this study I used spatial and temporal replication to document the patterns of predation in a fragmented Englemann Spruce/Subablpine Fir (ESSF) forest. This forest type occurs throughout the interior of British Columbia (Rowe 1972) and the northwestern United States, and is heavily logged across much of this range. The primary focus of my study was to examine the frequency of nest predation in forest edge, forest interior, and clearcut habitats. I tested the prediction that nest predation in clearcut and forest edge habitats would be greater than that documented in the forest interior. Due to the difficulty in finding an adequate sample of natural nests in forested habitats, I investigated patterns of nest predation in the ESSF forest using artificial nests. Although these nests lack important characteristics of natural nests, e.g. the incubating adults, they provide an index of relative predation risk. Artificial nests also allow control of sample size, nest placement, and nest characteristics (Martin 1988, Moller 1987, Storch 1991). A full understanding of the effects of forest fragmentation on nest predation also requires a knowledge of the habitat use by potential nest predators. Previous studies of the impact of fragmentation on open-nesting passerines have suggested that the following mechanisms explain the increase in nest predation at the forest edge: 1) an increase in the number of nest predators at the forest edge (Whitcomb et al. 1981, Wilcove 1985, Robinson et al, 1991); 2) a change in the species of predators at the forest edge (Andren and Angelstam 1988, Andren 1992); and 3) an increase in the activity and/or foraging success of predators at the forest edge (Ladino 1980, Gates and Gysel 1978, Gibbs 1991). However, few studies of nest predation have generated enough information to test these three hypotheses (Andren 1992, Angelstam 1986, Rosenberg and Rahael 1986). In this study, I documented the distribution and relative abundance of predators in the ESSF forest, and used this information to test the first two hypotheses. As I was not able to identify and follow individual predators, I did not study the foraging success of predators in clearcuts and at the forest edge. The thesis is organized into two sections. I first examine the pattern of nest predation in forest edge, forest interior, and clearcut habitats in the ESSF forest. I then document the distribution and identity of nest predators in these habitats. M E T H O D S Site Description This study was conducted in the ESSF forest region of the interior of British Columbia. Study plots were in a 256 km 2 area, centered 18 kilometers east of the town of Okanagan Falls, (49° 20' N, 119° 20' W), at an elevation of 1550-1700 meters (Figure 1, p. 30). The area is a mosaic of forest stands and clearcuts ranging in age from <1 years to >80 years old. The dominant canopy species in the region are Englemann Spruce (Picea englemanni), and Subalpine fir (Abies lasiocarpa), with Lodgepole Pine (Pinus contorta) in the drier upland and south-facing areas. Understory species included such low-growing shrubs as Grouseberry (Vaccinium scoparium), and Mountain Labrador Tea (Ledum glandulosum), as well as canopy species saplings. Clearcut areas were characterized by Bearberry (Arcostaphylos uva-ursi) and young Lodgepole pine, with small clumps of mature aspen (Populus tremuloides) in the wetter areas. Table 1 provides a summary of the height and distribution of the different vegetation layers in this forest. Eleven study plots were used over the two years of this study, 1993 and 1994. Potential plots were first located on 1:15 000 forest cover maps and then visited to assess similarity and ease of access. I chose plots on the basis of their similarity in species composition, slope (<20°), and aspect (north-facing slopes). Each plot 4 consisted of a clearcut (14-30 ha in size) ranging in age from 2-14 years, with no less than 500 meters of uninterrupted forest extending from at least one side. A l l study plots were at least 1 km apart. Patterns of Nest Predation in Forest Edge, Forest Interior, and Clearcut Habitats: Field Procedure (1993) To generate an index of nest predation, I placed artificial nests in forest edge and forest interior habitats at six different study plots (as per Seitz and Zegers 1994). Artificial nests were constructed using dried grasses collected from plots. Each nest was approximately 10 cm in diameter and 8 cm deep. Nest construction and placement were modeled after nests of common passerines nesting in the area, i.e. hermit thrush (Catharus guttatus), varied thrush (Ixoreus naevius) (shrub nests), and dark-eyed junco (Junco hyemalis) (ground nests). A summary of other passerine species nesting in this ESSF forest is presented in Appendix 1. At each plot parallel transects of nests were established in the forest along its edge ('edge' was defined as the point at which the forest canopy ended and the adjacent clearcut began (Ranney et al. 1981)), and 200m into the forest. Thirty meters separated each nest site along these transects, with flagging tape marking each interval. Nests were placed 5-15 m from the flags at the nearest suitable site. Field notes describing the location of each nest allowed them to be relocated on each visit. Nests were placed alternately along each transect on the ground or in a shrub. I placed ground nests in well concealed locations beneath a log or rock; shrub nests were placed in concealed positions in a shrub or young tree. The shrub nests were secured by 20 gauge galvanized wire to the central stem or trunk to reduce loss of contents due to tipping by predators and/or wind. Forty-eight nests were placed on each plot, with 24 (12 ground and 12 shrub) nests along each of the edge and interior transects. The experiments were conducted over a three month period, roughly 5 corresponding to the breeding season of local passerines (Cannings et al. 1987). Three trials were conducted in 1993: May 25-June 3 (Trial 1), June 16-July 1 (Trial 2), and July 6-July 17 (Trial 3), with approximately three weeks separating each trial. At the beginning of each trial, artificial nests were placed along the transects. In subsequent trials nests were moved at least 10 meters from their previous location. Each nest was baited with two quail eggs (Coturnix coturnix) and a plasticine egg. A l l nests and eggs were handled with latex gloves, and experimenters wore rubber boots in order to minimize human scent trails. To test whether plasticine eggs affected the likelihood that a nest would be depredated, alternate nests were given a plasticine egg in Trial 2 (1993). Nests were then left for a seven day exposure period; this period was chosen to facilitate comparison with other artificial nest studies. In Trial 1 only, nests were also checked after four days. This procedure was not repeated in subsequent trials to minimize potential observer impact on vegetation. Nests were checked for signs of disturbance after the appropriate exposure period. Any nest with >1 egg missing was considered depredated. Predation rate was calculated as the number of nests depredated divided by the total number of nests on each transect. Field Procedure (1994) In the second year of my study, I modified the experimental design to introduce clearcuts as an additional habitat treatment, and the interval between nests along all transects was widened from 30 to 50 meters to decrease the density of nests. Five new study plots were located, in addition to four used in 1993. Two trials were carried out in 1994: June 7-June 18 (Trial 1) and August 10-August 22 (Trial 2), with 6 weeks separating the beginning of each trial. Extensive nearby forest fires in 1994 caused the longer gap between trials. In Trial 1 (1994), ground and shrub nests were located as described for the trials in 1993. In Trial 2 (1994), I used only ground nests, but with two concealment 6 categories: high cover (75-100% concealed) and low cover (0-25% concealed). Concealment was scored while standing 2 m from the nest at the time of placement. For each trial, nests were baited with two quail eggs, and alternate nests were baited with a plasticine egg. The procedure for placing and checking the nests was the same as in 1993. A minimum of 28 nests were placed at each plot, with 14 nests (Trial 1: alternating ground and shrub, Trial 2: alternating high and low concealment) along each of the edge and interior transects. In six of the nine plots, a third nest transect was added in the clearcut. This subset of plots was chosen because they contained clearcuts which were large enough to allow the placement of transects of artificial nests, e.g. greater than 300 meters in length, and at least 150 m wide. The clearcut transects were parallel to the existing forest edge and forest interior transects, and were a minimum of 50 m from the forest edge. For Trial 1, ground nests were used in the clearcut (high cover), for Trial 2 both high cover and low cover nests were used in the clearcut. Predation rates were calculated as in 1993. Table 2 summarizes the experimental design for the two years of this study. The Distribution and Identity of Nest Predators in the ESSF Forest: Predator Identity (1993, 1994) In both 1993 and 1994,1 collected egg remains (both plasticine and quail eggs) from each nest at the end of a trial. Quai l egg remains were assigned to a mammalian or avian predator according to marks left on the shell and/or the shape of the shell remains. Mammals left obvious teeth marks on the exterior of the eggs, and/or the shell was chewed into small fragments (Trevor et al. 1991, Rearden 1951). A v i a n predators often pierced the shells to reach the contents of the quail eggs, leaving a large hole in an otherwise intact shell (Rearden 1951). I was able to identify mammals to the family level by the shape of teeth marks left on the plasticine eggs. Sample 'bites' on plasticine eggs for species known 7 to inhabit the ESSF forest were made from the skull collection at the Vertebrate Museum, University of British Columbia. Plasticine eggs from the field were then compared to these specimens. I was not able to identify avian predators to as detailed a level as mammals as the bill markings were indistinguishable among the possible candidates. In 1994 I conducted predator surveys on all the plots. Auditory and visual encounters with red squirrels (Tamiasciurus hudsonicus), gray jays (Perisoreus canadensis), and yellow-pine chipmunks (Eutamias amoenus) were recorded along each edge and interior nest transect. These surveys were conducted as a part of every visit to the transects over the course of the summer, and the average number of predators detected per plot was calculated from these values. Statistical Analysis: Predation Data I used three-way analyses of variance (ANOVA) to examine the effects of 'Plot' (the 5-9 forest plots), 'Location' (edge vs interior vs clearcut) and 'Height' (ground vs shrub) or 'Cover' (high vs low) on percent of nests depredated. 'Location', 'Height', and 'Cover' were treated as fixed factors, and 'Plot' was a random factor. Since this was a mixed-model A N O V A , with no replication for the factor 'Plot', F-statistics were calculated according to Table 3. However, where interactions were not significant (p>0.05), I pooled them with the error, and the main effects were tested over this new error variance. Although Zar (1984) recommends a conservative approach to pooling, I used pooled and non-pooled estimates of the error variance in my study, and found no difference in the results. The data were arcsine transformed to ensure that it met the assumptions of normality for an A N O V A . A Three-way A N O V A with repeated measures was used to examine whether nest predation rates in 1993 varied according to: 1) the habitat treatment effects, e.g. 8 'Plot' and 'Location', 2) time, e.g. the different trials, and 3) the interaction between these variables (see Bowers et al. (1993) for a similar use of repeated-measures ANOVAs ) . I tested for a difference in the frequency of predation between nest checks on Day 4 and 7 (Trial 1) 1993, and between Trials 2 and 3,1993. Due to missing data between Trial 1 and subsequent trials, I was unable to perform a repeated measures A N O V A for all trials simultaneously. Predator Data I used ANOVA ' s to examine the effect of 'Plot', 'Location', and 'Height' on the frequency of avian or mammalian predation in 1993 and 1994. As with the predation data, repeated-measures ANOVA ' s were used to examine the effect of 'Nest Check' or 'Trial' on the frequency of nests lost to avian and mammalian predators in 1993. These frequencies were calculated by dividing the number of nests lost to avian (or mammalian) predators by the total number of nests placed in each category. The data was arcsine transformed to ensure that it met the assumptions of normality for the A N O V A . A l l A N O V A s were performed using the General Linear Model procedure in SAS (SAS Institute Inc. Version 6.03), and JMP (SAS Institute Inc. 1994). Numbers of gray jays, red squirrels, and yellow-pine chipmunks estimated by auditory and visual surveys, were compared at edge and interior locations using Mann-Whitney 'U' tests. These analyses were performed using JMP (SAS Institute Inc. 1994), and SYSTAT (Systat Inc. 1989) computer packages. RESULTS Patterns of Nest Predation in Forest Edge, Forest Interior, and Clearcut Habitats: The effect of 'Plot' and 'Time' 'Plot' was a significant treatment effect in this study in both 1993 and 1994, with the frequency of predation varying among the different study plots. Predation rates varied across plots in Trial 1 (Table 4), and in Trials 2 and 3 (Table 5), 1993. In Trial 1 there was almost twice as much predation at plot 271(3) compared to the other plots (Figure 2), and this was consistent for both nest checks. In Trials 2 and 3, plots could be grouped according to those with an average frequency of 45%, and those with an average of 65% of nests depredated (Figure 3). 1 0 0 ' z o Q 5<H 6 0 ' 4 0 H Nest Check: Day 4 a 1 Nest Check: Day 7 FIGURE 2: The effect of 'Plot' on frequency of nest predation for nest checks on Day 4 and 7, Trial 1, 1993. FIGURE 3: The effect of the 'Plot'*'Trial' interaction on frequency of nest predation for Trials 2 and 3, 1993. 10 There was evidence that much of the variation between plots in 1993 and 1994 may be attributed to the sequence in which these plots were used, e.g. the frequency of nest predation increased significantly with time both within and between trials. Indeed, there was a significant increase in the average frequency of nest predation between nest checks in Trial 1 (Figure 2, Table 4), from 48+6% to 72+6% (data reported as mean+lse). The significant interaction of 'Plot' and 'Trial' in Trials 2 and 3 (Table 5) occurred because there was no difference in the frequency of predation between the two trials at some plots, e.g. 36(8), while at others there was a two-fold increase in nest predation, e.g. 231(10) (Figure 3). Plots which had been used previously in Trial 1, e.g. 36(8) and 59(14), did not exhibit an increase in predation between Trials 2 and 3. Conversely, plots which were first used in Trial 2, e.g. 231(10), exhibited an increase in predation between Trials 2 and 3. Plot to plot variation in percent of nests depredated was significant for the first trial in 1994 (Table 6), with a pattern of predation similar to that recorded in 1993 (Figure 4), e.g. some plots had high predation rates, with an average of 61% of nests depredated, while low predation plots lost an average of 34% of nests. Overall, plots which had been used previously in 1993, e.g. 59(14) and 231(10), had a higher predation rate than those which were used initially in 1994, e.g. 36(10) and 22(L). In the second trial all plots had a predation rate close to the average of 42% (Figure 4), and the overall predation rate was not significantly different from Trial 1 (t=0.65, p=0.5) In summary, there was significant variation in the average frequency of predation across the different study plots in both 1993 and 1994, and much of this variation may be attributed to the sequence in which the study plots were used. 100 • 5 0 4 2 60 H o 9 40 H 20 H z 0 1 T 1 DAY 7 T 1 Trial 1 T 1 T 1 o 1 1 1 1 1 1 1 1 r 36 (8) 59 (14) 59 (17) 231 (10) 59 (16) 36 (10) 23 (A) 22 (L) 68 (F) P L O T 100' 8 0 H 60 H 40 4 20-DAY 7 Trial 2 T T i ± T T 1 1 1 1 n 1— 59 (14) 59 (17) 231 (10) 23 (A) 22 (L) PLOT T 1 62 (F) FIGURE 4: Average predation rate for plots, Trials 1 and 2, 1994. Error bars indicate+lse, n=4. The effect of nest location Despite the significant temporal variation in nest predation rates, I found a difference in the frequency of predation between nest locations, e.g. interior, edge, and clearcut, in both 1993 and 1994. Nest location was a significant between-subject; effect in Trials 2 and 3, 1993 (Table 5); yet, the pattern of nest predation at edge and interior locations in these trials was opposite to that documented in previous studies. The average frequency of predation in the forest interior (62+3.6%), was slightly higher than at the forest edge (55+3.9%). There was no difference in the frequency of predation at edge and interior nest locations in Trial 1, 1993 (Table 4). The significant effect of nest location in Trial 2,1994 (Table 7), was because nests in clearcuts had a significantly lower frequency of predation than either edge or interior nests (one-tailed t-tests: 1=3.81, df=22, p=0.001, and t=3.56, df=22, p=0.002 respectively) (Figure 5). There was no difference in the frequency of nest predation between the two forest nest locations (edge and interior) in this trial (one-tailed t-test, t=1.01, df=22, p=0.33). In Trial 1 the frequency of predation in clearcuts was also lower than either forest locations, but this difference was not significant (Figure 5, Table 7). 12 z o p < Q § a. CLEARCUT EDGE INTERIOR Z o p < Q CLEARCUT EDGE L O C A T I O N INTERIOR FIGURE 5: The effect of Location' on frequency of nest predation for Trials 1 and 2, 1994. Error bars indicate ±lse, n=6. 13 In Trial 1, 1994, the interaction between 'Plot' and 'Location' was significant (Table 6), e.g. there was significant variation among plots in the difference in the frequency of nest predation between edge and interior locations (Figure 6). At some plots, predation on nests at the edge was much greater than predation in the forest interior, e.g. plot 59(16), yet at other plots there was little difference in the predation rate at edge and interior locations, e.g. plot 22 (L) (Figure 6). Overall, nest predation at the edge was significantly higher than predation in the interior (one-tailed t-test, t=1.87, df=17, p<0.05), this pattern is opposite to that recorded for Trial 2,1994, or Trials 2 and 3, 1993. 100' z o p < Q 75 A 5 0 H 25 A X A A X A X T 1 1 1 ¥ 1 1 1— 36 (8) 59 (14) 59 (17) 231 (10) 59 (16) 36 (10) 23 (A) 22 (L) 62 (F) A X A EDGE X INTERIOR X A FIGURE 6: The effect of the interaction of 'Location' (edge vs interior) and 'Plot' (in the Three-way ANOVA) on frequency of nest predation for Trial 1, 1994. The effect of nest height and cover In many cases, predation on shrub nests was greater than predation on ground nests, although these differences were significant only for Trial 1, 1994 (Table 6). In this trial, the average frequency of nest predation on ground nests was 40±7%, while 53+8% of shrub nests were lost. Although shrub nests were often less concealed than ground nests, there was no indication that nest cover alone 14 influenced predation rates (Table 7). In Trial 2, 1994, nests with less than 25% concealment were no more likely to be depredated than nests wi th greater than 75% concealment. The effect of nest contents In both years of study, predation on nests with plasticine and quail eggs was higher than in nests with quail eggs only (1993: one-tailed t-test, t=-4.36, df= 11, p=0.0001; 1994: t=-3.74, df=23, p=0.001). In 1993 nests without plasticine were depredated at a rate of 40+5%, while nests with plasticine lost 67+3% nests. The pattern was almost identical for 1994: nests without plasticine had an average loss of 43+5%, while the predation rate on nests with plasticine was 67+5%. In summary, predation on artificial nests in the ESSF forest was affected by: 1) the location of the nest, e.g. nests in clearcuts were less likely to be depredated than nests in the forest, 2) the sequence in which study plots were used, and 3) the contents of the nest (plasticine vs quail eggs). There was little evidence that nest height or cover were important determinants of the risk of nest predation. The Distribution and Identity of Nest Predators in the ESSF Forest: Overall Patterns Plasticine and quail eggs allowed me to identify the predator at an average of 36% of depredated nests. The remainder of nests either had no egg remains (50%), or contained remains that I could not attribute to a specific predator (14%). The most common mammalian predators identified from eggs in the forest were red squirrels, yellow-pine chipmunks, and deer mice (See Appendix 2 for full list and Latin names). Gray jays, American crows, and common ravens were the most prevalent potential avian predators in the ESSF forest, although I could not identify avian species from the beak marks remaining on the eggs. Deermice, chipmunks, and an avian predator were most commonly identified from egg remains in 15 clearcuts, while red squirrel predation was seldom recorded in the clearcut. In both 1993 and 1994, there were more cases of mammalian than avian predation, but the ratio of avian/mammalian predators was not constant across trials. This ratio increased from 1:8 in Trial 1 (four day nest check) to 1:1.5 in Trial 3, 1993 (Figure 7). In 1994, the ratio of avian to mammalian predation was 1:3 in Trial 1 and 1:5 in Trial 2 (Figure 7), and the ratios were similar for both clearcut and forest nest locations. I suspect that the low number of avian predators in Trial 2, 1994 was because I only used ground nests in this trial. 2 5 • 2 0 H 1 5 NUMBER OF PREDATORS DETECTED 1 0 H 5 H 1:8 1:3 1:2 1— 1 r 1 (DAY 4) 1 (DAY 7) 2 1:1.5 EJ3 MAMMALIAN • AVIAN 1:5 —I— 1 (1994) 2 (1994) TRIAL FIGURE 7: Average number of avian and mammalian predators detected at study plots, Trials 1-3, 1993, and Trials 1-2, 1994. Numbers above bars indicate ratio of avian to mammalian predators. Temporal and Spatial Patterns There was an overall increase in the frequency of identified nest predators between Trials 2 and 3, 1993 (Table 9). The proportion of mammalian predation increased from 15+3% in Trial 2, to 29+4% in Trial 3, while the frequency of avian predation almost doubled, from 11+1% to 21+3%. There was no evidence for an 16 increase in the frequency of nest predation by mammalian or avian predators between nest checks in Trial 1, 1993 (Table 8). Despite considerable variation in the frequency of nest predation among the different study plots in 1993 and 1994, the frequency of mammalian and avian predation was not significantly different among these same plots (Table 8, 9b and 10). Only in Trials 2 and 3 was there a significant effect of 'Plot' on the proportion of nests lost to mammalian predators (Table 9a). Mammalian predation ranged from 12% to 31% (plots 59(14) and 231(10) respectively) (Figure 8). However, the frequency of nests lost to mammalian predators was not related to the overall nest predation rate at these study plots. Plots with a high frequency of nest predation, e.g. plots 59(17) and 231(10), d id not have a correspondingly high frequency of nests lost to mammalian predators, while the plot wi th the lowest proportion of mammalian predation, 59(14), had a high average nest predation rate (Figure 8). z ^  P < P P 36(8) 68(30) FIGURE 8: Average frequency of nest predation attributed to mammalian predators at each study plot in Trials 2 and 3. Filled circles indicate the average frequency of predation recorded at these plots. Error bars indicate +lse, n=4. There was no difference in the frequency of avian and mammalian predation at edge and interior nest locations in either year (Tables 8-10). However, avian and 17 mammalian predators preferentially preyed on nests of different heights. In 1993, mammals were consistently more likely to prey on ground nests; while avian predation was more commonly associated with shrub nests (Tables 8-9, Figure 9). In 1994, mammalian predators preyed on ground nests twice as often as shrub nests, but avian predators were equally likely to attack ground and shrub nests (Table 10, Figure 9). 4 0 -3 0 -z o B 2 0 -Q I ioH 1 * T 1 T T I 1 1 12] GROUND NEST • SHRUB NEST 1 T j _ 1 MAMMALIAN AVIAN T r i a l 1, 1993 MAMMALIAN AVIAN Trials 2 and 3, 1993 MAMMALIAN AVIAN T r i a l 1, 1994 FIGURE 9: Average frequency of predation attributed to mammalian and avian predators at ground and shrub nests. Bars indicate+lse. Trial 1, 1993: n=3; Trials 2 and 3, 1993: n=5; Trial 1, 1994: n=9. '*' indicates a significant difference, p<0.05. Nest Contents Overall, I was able to identify over twice as many predators with plasticine eggs as compared to quail eggs. However, the ratio of avian to mammalian predators varied with the contents of the nests. In 1993, nests which contained only quail eggs had a ratio of 1 avian predator to 1.1 mammalian predators, while nests with plasticine eggs had a ratio of 1:2.5. The pattern was similar in 1994, nests with quail eggs had an avian to mammalian predator ratio of 1:1.6, while nests with plasticine had a ratio of 1:3.3. 18 The most consistent pattern of predator identity in this study was an increase in the ratio of avian to mammalian predators across trials, and an increase in the number of mammals identified at nests which contained plasticine eggs. While there was a significant difference in the height at which avian and mammalian predators foraged, there was no difference in the frequency of avian or mammalian predation associated with the forest edge, forest interior or clearcut. Patterns of Predator Distribution Across Habitats from Auditory and Visual Surveys (1994 only) The visual and auditory surveys of potential nest predators revealed differences in their relative abundance. A n average of 6 red squirrels, 1-2 chipmunks, and 1-2 gray jays were recorded at each study plot. However, the number of these predators identified per plot d id not correlate wi th predation rates at those same plots (red squirrels: r=-0.56, p=0.12; chipmunks: r=0.12, p=0.4; gray jays: r=0.01, p=0.9). Sites with high predation, e.g. 59(14), d id not have correspondingly large numbers of total predator detections, and sites wi th low frequencies of predation, e.g. 22(L), often had the greatest number of predator detections overall. Audi tory and visual surveys revealed no difference in the number of red squirrels detected along edge and interior transects (Mann-Whitney U test, p>0.2) (Figure 10). Chipmunks and gray jays were more common at the forest edge than in the forest interior, although this difference was not significant (Mann-Whitney U tests, p>0.2) (Figure 10). Although there were slightly more gray jays and chipmunks at the forest edge, this d id not translate to either increased predation at the forest edge or to an increase in the number of these predators identified from plasticine eggs. 19 3 H NUMBER OF INDIVIDUALS 2 DETECTED 1 H E3 EDGE • INTERIOR T R E D S Q U I R R E L C H I P M U N K PREDATOR G R A Y J A Y FIGURE 10: Average number of individual predators detected during auditory and visual surveys of edge and interior transects, n=9. Overall, the distributions of potential nest predators at the forest edge and forest interior were similar, and did not reflect the decreased frequency of nest predation at the forest edge. Although there was variation in the number of predators detected among the different study plots, this did not correspond with the patterns of nest predation at those plots. Differences in predation rate between previously 'exposed' and 'unexposed' plots: a post-hoc test As was discussed earlier, significant temporal variation in predation rates was apparent both within and between trials. I therefore tested if this variation was related to whether the study plot had been previously exposed to artificial nests, or not. In Trial 2,1993, three plots were 'carried over' from Trial 1 (Plots: 36(8), 59(14), and 271(3)), and three new plots were added (Plots: 59(17), 231(10), and 62(F)). The difference in predation rate between these two groups of plots was significant (one tailed t-test, t=3.84, df=22, p=0.0001). The average predation rate for the three previously exposed plots was 63+3.6%, higher than the average predation rate for the previously unexposed plots (45+0.7%). The number of nest predators identified 20 was also different. At 'exposed' sites 32 predators were identified, with a ratio of 1 avian predator to 2 mammalian predators; 22 predators were identified at 'unexposed' sites, with a ratio of 1:5 (avian to mammalian). This pattern was also apparent between years. Four plots from 1993 were used again in 1994; these previously exposed plots had a higher predation rate than the five new plots, 55±4.5% and 39±7.2% respectively (one tailed t-test, t=2.68, df=34, p=0.08). This comparison may be confounded because three of the five new plots in 1994 were approximately 10 km from the majority of the other plots (Figure 1). The number of predators identified from plasticine eggs, however, did not reflect this pattern, with an average of 5 predators identified at each site, regardless of exposure. Nonetheless, the ratio of avian predators to mammalian predators was greater for 'exposed' sites (1:1.5) than at 'unexposed' sites (1:4). In summary, much of the variation in predation rates both within and between trials was due to the sequence in which study plots were used. Plots which had been previously exposed to my artificial nests had higher predation rates, a greater number of predators identified from plasticine eggs, as well as a greater number of avian predators (relative to mammalian predators). DISCUSSION Patterns of Nest Predation: The Effect of Fragmentation In this study, I found no evidence to support the prediction that nest predation at the forest edge is greater than in the undisturbed forest interior. This is in contrast to the findings of Gates and Gysel (1978), Andren et al. (1985), and Wilcove et al. (1986), who found an increase in predation on nests located at the forest edge. However, these studies were conducted in agriculturally-dominated landscapes, with extensive human development in the surrounding habitat matrix. Other research in forest-dominated landscapes has failed to find a difference in the 21 predation rates between edge and interior nest locations (Rudnicky and Hunter 1993, Small and Hunter 1988). Storch (1991) suggested that edge effects may only occur in areas where the productivity of the surrounding matrix supports greater predator densities than the habitat it surrounds, e.g. an agricultural or urban setting. The network of roads and clearcuts which fragment the Englemann Spruce/Subalpine Fir forest may not support, or facilitate foraging by, a large community of potential nest predators, and thus not generate the edge effects documented in previous studies. Indeed, fragmentation may lower the densities and/or foraging activity of key predators, e.g. gray jays, red squirrels, and deer mice. Storch's hypothesis is supported by the low frequency of predation in clearcuts I documented. Thus, the frequency of nest predation in the forest interior may reflect levels normally encountered within this forest. I present a simple graphical model (Figure 11) to describe the relative predation risk within clearcut, edge, and interior habitats in the ESSF forest. Future studies should establish the frequency of nest predation at intermediate distances from the edge. z o H < Q C L E A R C U T E D G E FOREST FIGURE 11: Graphical model of relative frequency of predation on artificial nests in clearcut, forest edge, and forest interior habitats. Although I found decreased predation in clearcuts wi th my artificial nests; 22 predator surveys revealed no significant differences in their distribution across forest and clearcut habitats. Indeed, a live-trapping study of small mammals on the same plots (Blomer 1995), revealed a slight increase in the number of deermice and chipmunks in clearcuts. Thus the data does not fully support Storch's prediction that the clearcut matrix harbours a smaller number of predators than the forest. More research should be directed at documenting predator abundance and activity within clearcuts, and how these changes might influence the remaining forest habitat. The low frequency of predation in clearcuts may be more a result of the lack of overhead canopy, than of intrinsic differences in predator numbers. Predators may be less likely to spend time searching for nests in a clearcut as they may be vulnerable to aerial attacks themselves, e.g. I observed red-tailed hawk (Buteo jamaicensis) predation on a ground squirrel (Citellus columbianus), and a Merlin (Falco columbarius) pursue a gray jay in a clearcut. Rudnicky and Hunter (1993) found that clearcut nests with greater cover were more likely to be depredated than more exposed nests. Similarly, Santos and Telleria (1992) documented a decrease in predation at the edge of their forest plots. They attributed this decrease to the restricted activity of small forest predators in response to the increased incidence of badger and fox predation in the surrounding landscape. In a situation in which no overhead canopy is present; predators may then preferentially prey on nests surrounded by greater vegetation cover. In a forest, where the canopy provides protection from aerial predators, cover at the nest may not be an important component of predation risk. Many studies have tested the hypothesis that foraging behaviour is influenced by predation risk, e.g. Bowers and Ellis (1993), Bowers et al. (1993) . In these studies chipmunks and squirrels were presented with food sources in areas of high and low vegetation cover (predation risk was assumed to be inversely correlated with cover), and study animals exploited more seeds under high cover. It 23 would be interesting to similarly document the foraging activity of individual predators in clearcut, edge and forest interior habitats. Despite the fact that there was a decrease in the frequency of nest predation in clearcuts in my study, there were no consistent differences in the nest predation rate at edge and interior locations. Documented predation rates at these locations were often contradictory, e.g. in Trials 2 and 3,1993, predation at the forest edge was slightly higher than that in the forest interior, while in Trial 1, 1994, the frequency of predation at the forest edge was often greater than in the forest interior. A possible explanation for this lack of a consistent difference between edge and interior nest predation rates may be the habitat use of potential nest predators at each of my study plots. The principle nest predators identified in my study, e.g. gray jays, red squirrels and chipmunks may not alter their foraging activities at the forest edge. There was little difference in their distribution across the different forest locations as documented by my artificial nests and auditory and visual surveys, as well as by a live-trapping study at these same forest locations (Blomer 1995). This pattern of habitat use may be confounded by the fact that many of these predators have home ranges large enough to encompass a number of my study plots, e.g. gray jays and common ravens (Shank 1986). Thus much of the nest predation at a particular plot may be the result of individual predators moving through the area, rather than intrinsic differences between the forest edge and forest interior. In summary, fragmentation of the ESSF forest may result in decreased predation rates in clearcuts, while predation in the forest remains unchanged. Whereas nest predation in agricultural landscapes may be elevated because of greater number of nest predators in the agricultural land, nest predation in forests fragmented by logging activity may be decreased because of the increased risk to small mammals and birds foraging in clearcuts and/or at the forest edge. Despite the decrease in nest predation in clearcuts, there was no consistent difference in the frequency of predation at edge and interior nest locations throughout the ESSF 24 forest. This pattern may be attributed to differences in the foraging strategies and habitat use of potential nest predators. While small mammals may avoid foraging in the open clearcuts, their use of the forest edge and the forest interior does not differ. Patterns of Nest Predation: The Effect of Nest Height and Cover The influence of cover on predation rate was complex. Shrub nests, which were often poorly concealed, were preyed on more often than ground nests (see Table 5), yet I was unable to detect an influence of cover on predation rate within ground nests (see Table 6). Seitz and Zegers (1993) suggest that ground-nest survivorship is higher than for shrub nests because of the lack of well-concealed nest sites in many coniferous forests. In my study forest the understory was open, and shrub nests may have been more vulnerable to predation because the vegetation in which they were located was itself conspicuous. Predators may have quickly learned to associate my artificial nests with small conifers or clumps of alder. Indeed, Martin and Roper (1988) found that hermit thrush nests were preyed on less often if they were located in areas with many potential (but unused) nesting sites. Despite concerns about mimicking natural nest sites, the patterns of predation on artificial ground and shrub nests documented in this study are similar to those found for natural nests in a similar forest type (Martin 1993). However, Wilcove (1985) found that artificial ground nests were more vulnerable than shrub nests in the Eastern U.S.A., and that this corresponded with the nesting habits of many of the species sensitive to habitat fragmentation, e.g. the ovenbird and hooded warbler (Wilsonia citrina). It remains to be seen whether shrub-nesting birds in the ESSF forest suffer higher predation rates, and whether they are more likely to suffer declines, as this forest is fragmented. Study of the patterns of predation on natural nests, e.g. varied thrush (shrub nester) and orange-crowned warbler (ground nester), 25 will allow us to establish the vulnerability of these different nesting guilds within this fragmented ESSF forest. Patterns of Nest Predation: Temporal Effects In both 1993 and 1994, there was an interesting difference in the frequency of predation on exposed and unexposed plots; plots which had been previously exposed to my artificial nests had a higher frequency of predation than those which were new to the study. This difference may be explained by the positive reinforcement of predators on previous trips to these plots. Predators routinely spend more time foraging in areas where they have been successful (Stephens and Krebs 1986, Sugden and Beyersbergen 1986). Both avian and mammalian predators will return to nests that they have found previously (Willebrand and Marcstrom 1988, Sonerud 1985), and repeat visits may occur both within and between years. Indeed, Sonerud and Fjeld (1987) found that artificial nests that were depredated in the previous year, were significantly more likely to be preyed upon than nests placed in new locations. In my own study, predators may have retained a memory of my nests/or nest areas and thus have been able to exploit this food source more efficiently in subsequent trials. This may be particularly true for avian predators such as corvids, including gray jays, which rely on spatial memory to recover food cached for the winter (Bunch and Tomback 1986). However, the between-year differences in predation rate between exposed and unexposed nests must be interpreted with caution, as this pattern of exposure may be confounded by the general location of plots within the landscape. Many of the plots which were first used in 1994 were at least 10 km from the majority of the other plots (Figure 1). Temporal shifts in the frequency of nest predation were a source of variation in this study. A n increased frequency of predation both within and between years 26 may have been the result of a concomitant increase in the ability of potential predators to exploit my nests. While this may have been an artifact of the methodology, it corresponds with previous findings that natural nests are vulnerable to repeat visits from nest predators (Sonerud 1985). Artificial nests and plasticine eggs: how good is the evidence? I was able to identify 20-56% of predators from plasticine eggs. In a similar study Nour et al (1993) were able to identify 50% of predation events from plasticine eggs in a fragmented Belgian forest. Studies using alternative methods of identifying predators were able to identify a greater proportion of predators visiting artificial nests. Leimgruber et al. (1994) were able to identify 67% of predators at nests using cameras, while Angelstam (1986) used 'grease boards' at artificial nests to record 75% of the predators which visited his nests. However, the amount of information derived from these three methods must be weighed against three factors: 1) their expense (high for cameras, low for 'grease boards' and plasticine eggs), 2) the ease of implementation, and 3) their inherent biases. Plasticine eggs provide an inexpensive and convenient alternative to either of these methods outlined above, and produce only slightly lower rates of predator identification. In general, there is controversy on the quality of evidence derived from artificial nests. Leimgruber et al. (1994) found that artificial nests were often visited by predators, but that the eggs were not touched. Thus, artificial nests may underestimate the visitation rate of predators, particularly if the predators are incapable of handling quail eggs. Mice or other small predators may not be able to gnaw their way through the tough shell of the quail egg, as they would with a smaller, softer passerine egg (Boag et al. 1984). Conversely, artificial nests may be biased to a certain type of predator, Willebrand and Marcstrom (1988) suggested that nests with 'grease boards' may be conspicuous to visually searching predators, thus overestimating their abundance. Similarly, plasticine eggs may overestimate the relative frequency of predation by small mammals, and hence their importance, as nest predators. Small mammals may be more likely to 'handle' the plasticine egg, and thus leave a recognizable mark, than the quail eggs. Equally likely, small mammals may be attracted to the smell of the plasticine, and be better able to exploit a series of artificial nests. These hypotheses are supported by the fact that nests with plasticine eggs were more likely to be depredated, and had a greater number of mammalian predators, than nests with just quail eggs. M y study revealed other problems associated with this methodology. I was often unable to identify the predator(s) responsible for egg losses, particularly if nests contained only quail eggs. There may have been a predator which contributed greatly to the patterns of predation documented in this study, but which I failed to identify from the egg remains. Many study plots with a high frequency of nest predation did not have a correspondingly high number of identified predators. Indeed, Rearden (1951) and Trevor et al. (1991) suggest that predators such as American crows (Corvus brachyrhynchos) and red foxes (Vulpes fulva), may carry the nest contents away and consume them elsewhere. Thus I may have underestimated the importance of these predators because I was not able to detect their presence. A complementary study with remote cameras at a subset of artificial nests would help to establish the accuracy of plasticine eggs in identifying nest predators. Although artificial nests provide an index of predation in a number of different habitat types, we can only speculate whether artificial nests provide an unbiased index of nest predation rates in a variety of habitats, or whether they selectively identify a certain subset of predators. 28 Management implications and future consideration: Artificial nests have provided a simple means to document the patterns of nest predation in a variety of landscapes, and have helped to identify potential mechanisms operating in forest-bird communities, e.g. nest predation, Wilcove (1985). Artificial nests are a useful tool for nest predation studies conducted at a large temporal or landscape scale, because they can be standardized and replicated easily. However, we should recognize the limits of this method, and work to improve our understanding of the mechanisms of nest predation in both fragmented and pristine habitats. To further our understanding of the processes of nest predation in fragmented landscapes, future studies should address the following questions: 1) how does the foraging activity of potential nest predators differ between edge, interior, and clearcut habitats? 2) how do predators find nests, e.g. how does the concealment or spatial pattern of nests influence predation risk? Finally, 3) what is the predation rate on natural nests in these same habitats, and which predators are responsible for it? CONCLUSIONS Birds nesting within the Englemann Spruce/Subalpine Fir forest do not experience increased predation rates at forest edges. If anything, they may experience a decreased predation risk in clearcuts and, possibly, at the forest edge. Where this pattern is evident, it may not be the result of a decrease in the number or type of potential nest predators in this habitat, but rather of a decrease in the amount of time that predators spend foraging in clearcuts and at the forest edge. Decreased predation at the forest edge and in clearcuts may be found in other forest types that are fragmented by logging. However, we need to learn more about 29 the factors which influence nest predator foraging in clearcuts, at the forest edge, as well as in unexploited forests. While some forest types do harbour greater densities of potential nest predators in clearcuts, e.g. coastal Subalpine fir forests (Walters 1991), Douglas fir forests (Montana) (Ramirez and Hornocker 1981), and Appalachian oak forests (Buckner and Sure 1985), no studies have documented whether such increases translate to an increased nest predation risk in these same habitats. In addition, although nest predation rates may not be increased within forests fragmented by logging activities, there are other negative consequences of logging. Particularly, losses of mature stands of timber are likely to greatly reduce the numbers of species which rely on nest cavities, e.g. three toed wood-pecker (Picoides tridactylus), boreal owl (Aegolius funereus), and flying squirrel (Glaucomys sabrinus). Finally, studies using artificial nests involve several problems associated with this methodology. In my study there may have been biases associated with the type of predator documented by the plasticine eggs, and problems in identifying the predators. Future studies of nest predation in fragmented landscapes should compare natural with artificial nests, and use remote cameras at a subset of nests in order to establish the accuracy of plasticine eggs in identifying all nest predators. Only then will we be able to accurately document the mechanisms of nest predation in fragmented ecosystems. TABLE 1: Height and distribution of the different vegetation layers in the ESSF forest and associated clearcuts. Data are a summary of vegetation measurements and observations recorded for each study plot. V E G E T A T I O N LAYERS: FOREST HEIGHT SPACING INTERVAL Canopy 23-26m 3m Saplings 5-15m 5m Shrubs 0-2m 2-3m V E G E T A T I O N LAYERS: C L E A R C U T Saplings 0-2m 3-4 m, but patchy Shrubs 0-2m patchy cu 6 3 C CU ro cu £ 3 z ON CN T5 3 ro co* CA CA co (A 6 0 S it <u c c 6 l c 00 co i -o> cu 6 o co CO - . cu 00 00 03 cu CU o co rt* CN 00 rj* CO >-< cu cu E o co CO cu c ro CN rt* i 00 CN CU 4-* cu o CD CU C P CU CM • M i (C rt* t H CU re CN r}* i 00 CN CU cu o IT) CO cu CU CN re r}* Ui — cu 4-1 re H Z » © 8^  5 U 3 U z 52 oo C rr* 52 oo 52 oo C t^* CO , to CN 0) rt* 3 oo =5 ^ re co _ CO CN CU rt* 3 oo — CN, re ""^ CM CJ E SB H cn W Z CN II C J ^ N CM 11 CJ ^ CN ^ CM C J ^ T3 C 3 •—' t-^  i -3 3 . tool 3 u re cu U CU QJ O o U cj .SP 5 cu OH X 0) o re E E 3 CD CM w CD < Z o p < u o - J H cn tu Z cu . 2 too ui T3 CU w 3 CU . 2 bo C T 3 CU W "3 bo u i T3 CU cu . 2 T3 CU W 3 CO Ol 'co CA O CU too t J W co cu O^ 3 u i -ro cu 0 co co CA CA co CA ON CM UO CO CA CA CA r f CA CA S o 1- _ QJ VO 4-> ' ' =1 3 cj >-ro cu U VD r f CA CA CM TABLE 3: Source of variation and calculation of F-values for the mixed-model Three-way A N O V A . Factors 'Location' and 'Height' are fixed, factor 'Plot' is random (and not replicated). SOURCE OF VARIAT ION (Factors) Values for calculation of F-statistic Plot MSpiot/MSerror Location MSiocation/MSpiot*location Height/Cover MSheight/MSpiot*height Plot*Location MSpiot*locauon/MS e r ror Plot*Height MS piot*height/MS error Location*Height MSiocation*height/MSerror Plot*Location*Height/Cover Not possible to test, pooled with error variance TABLE 4: Results of the repeated measures A N O V A on frequency of nest predation between day 4 and 7 nest checks Trial 1,1993. Nonsignificant interactions (p>0.05) were pooled with the error. S O U R C E df MS F-value p-value Within Subject Effects Nest Check 1 3700 25.6 0.002 Error 7 115 Between Subject Effects Plot 2 1746 35.9 0.002 Location 1 60 1.2 0.3 Height 1 43 0.9 0.4 Error 7 49 TABLE 5: Results of the repeated measures A N O V A on frequency of nest predation between Trials 2 and 3, 1993. Nonsignificant interactions (p>0.05) were pooled with the error. S O U R C E df MS F-value p-vaiue Within Subject Effects Trial 1 1809 9.3 0.009 Trial*Plot 4 958 4.9 0.01 Error 13 194 Between Subject Effects Plot 4 359 6.5 0.004 Location 1 469 8.4 0.01 Height 1 180 3.3 0.09 Error 13 56 TABLE 6: Results of three-way A N O V A on frequency of nest predation in Trial 1, 1994. Nonsignificant interactions (p>0.05) were pooled with the error. S O U R C E df MS F-value p-value Plot 8 1307 11.9 <0.001 Location 1 1141 2.2 0.18 Height 1 1524 13.9 0.002 Plot*Location 8 605 5.5 0.002 Error 17 110 TABLE 7: Results of Three-way A N O V A s on frequency of nest predation, with 'Clearcut' as an added habitat treatment. Nonsignificant interactions (p>0.05) were pooled with the error. S O U R C E df MS F-value p-value TRIAL 1 (1994) Plot 5 1037 21.2 0.001 Location 2 209 0.6 0.46 Height 1 820 16.8 0.006 Plot x Location 10 330 6.7 0.01 Error 11 49 TRIAL 2 (1994) Plot 5 146 0.04 0.81 Location 2 " 2853 8.5 u 0.001 Cover 1 71 0.2 0.64 Error 27 372 TABLE 8: Results of the repeated measures A N O V A s on frequency of mammalian and avian predation on nests between day 4 and 7 nest checks Trial 1, 1993. Nonsignificant interactions (p>0.05) were pooled with the error. a: mammalian S O U R C E df MS F-value p-value Within Subject Effects Nest Check 1 63 0.5 0.51 Error 7 133 Between Subject Effects Plot 2 137 0.5 0.65 Location 1 15 0.05 0.83 Height 1 1134 3.9 0.09 Error 7 295 b: avian S O U R C E df MS F-value p-value Within Subject Effects Nest Check 1 100 1-4 0.28 Error 7 73 Between Subject Effects Plot 2 98 2.6 0.14 Location 1 5 0.1 0.72 Height 1 876 . 23.5 0.002 Error 7 37 TABLE 9: Results of the repeated measures A N O V A s on frequency of mammalian and avian predation on nests between Trials 2 and 3, 1993. Nonsignificant interactions (p>0.05) were pooled with the error. a: mammalian S O U R C E df MS F-value p-value Within Subject Effects Trial 1 1946 24 0.0003 Error 13 82 Between Subject Effects Plot 4 375 4.1 0.02 Location 1 51 0.6 0.47 Height 1 2481 27 0.0002 Error 13 92 b: avian df MS F-value p-value S O U R C E Within Subjects Effects 0.01 Trial 1 1145 8.8 Error 13 131 Between Subject Effects Plot 4 137 2.0 0.16 Location 1 78 1.1 0.31 Height 1 1416 20.5 0.0006 Error 13 69 TABLE 10: Results of Three-way A N O V A s on frequency of mammalian and avian predation on nests, with 'Clearcut' as an added habitat treatment, Trial 1, 1994. Nonsignificant interactions (p>0.05) were pooled with the error. a: mammalian S O U R C E df MS F-value p-value Plot 5 54 0.4 0.83 Location 2 56 0.4 0.65 Height 1 782 6.1 0.02 Error 21 128 b: avian S O U R C E df MS F-value p-value Plot 5 143 1.3 0.29 Location 2 196 1.8 0.19 Height 1 43 0.4 0.54 Error 21 107 REFERENCES 37 Andren, H . 1992. Corvid density and nest predation in relation to forest fragmentation: a landscape perspective. Ecology 73:794-804. Andren, H. and P. Angelstam. 1988. Elevated predation rates as an edge effect in habitat islands: experimental evidence. Ecology 69:544-547. Andren, H., P. Angelstam, E. Lindstrom and P. Widen. 1985. Differences in predation pressure in relation to habitat fragmentation: an experiment. Oikos 45:273-277. Anglestam, P. 1986. Predation on ground-nesting bird's nests in relation to predator densities and habitat edge. Oikos 47:365-373. Askins, R. 1995. Hostile landscapes and the decline of migratory songbirds. Science 267:1956-1957. Blomer, L. 1995. B.Sc. Honours' Thesis. Department of Zoology, University of British Columbia. Vancouver, BC. 22pp. Boag, D.A., S.G. Reebs and M.A. Schroeder. 1984. Egg loss among spruce grouse inhabiting lodgepole pine forests. Can. J. Zool. 62:1034-1037. Bowers, M.A. and A. Ellis. 1993. Load size variation in the eastern chipmunk, Tamias striatus: the importance of distance from burrow and canopy cover. Ethology 94:72-82. Bowers, M.A., J.L. Jefferson and M.G. Kuebler. 1993. Variation in giving-up densities of foraging chipmunks (Tamias striatus) and squirrels (Sciurus carolinensis). Oikos 66:229-236. Brittingham, M.C. and S.A. Temple. 1983. Have cowbirds caused forest songbirds to decline? Bioscience 33:31-35. Buckner, C A . and D.J. Shure. 1985. The response of Peromyscus to forest opening size in the southern Appalachian mountains. J. Mamm. 66:299-307. Bunch, K.G. and D.F. Tomback. 1986. Bolus recovery by gray jays: an experimental analysis. Anim. Behav. 34:754-762. Cannings, R.A., R.J. Cannings, and S.G. Cannings. 1987. Birds of the Okanagan Valley. British Columbia. The Royal British Columbia Musem, Victoria, BC. Gates, J.E. and L.W. Gysel. 1978. Avian nest dispersion and fledging success in field-38 forest ecotones. Ecology 59:871-883. Gibbs, J.P. 1991. Avian nest predation in a tropical wet forest: an experimental study. Oikos 60:155-161. Ladino, A. 1980. Animal activity patterns in transmission-line corridors and adjacent habitats. M.Sc. Thesis. Frostburg State College, MD. Leimgruber, P., W.J. McShea and J.H. Rappole. 1994. Predation on artificial nests in large forest blocks. J. Wildl. Manage. 58:254-260. Martin, T.E. 1993. Nest predation among vegetation layers and habitat types: revising the dogmas. Am. Nat. 141:897-913. Martin, T.E. 1988. On the advantage of being different: nest predation and the coexistence of bird species. Proc. Nat. Acad. Sci. U.S.A. 85:2196-2199. Martin, T.E. and J.J. Roper. 1988. Nest predation and nest-site selection of a western population of the hermit thrush. The Condor 90:51-57. Moller, A.P. 1987. Egg predation as a selective factor for nest design: an experiment. Oikos 50:91-94. Murcia, C. 1995. Edge effects in fragmented forests: implications for conservation. Trends. Ecol. Evol. 10:58-62. Norment, C.J. 1991. Bird use of forest patches in the subalpine forest-alpine tundra ecotone of the Beartooth Mountains, Wyoming. Northwest Science 65:1-9. Nour, N., E. Matthysen, and A.A. Dhondt. 1993. Artificial nest predation and habitat fragmentation: different trends in bird and mammal predators. Ecography 16:111-116. Paton, P. 1994. The effect of edge on avian nest success: how strong is the evidence? Conservation Biology 8(l):17-26. Ramirez, P. and M. Hornocker. 1981. Small mammal populations in different-aged clearcuts in northwestern Montana. J. Mamm. 62:400-403. Ranney, J.W., M.C. Bruner, J.B. Levinson. 1981. "The importance of edge in the structure and dynamics of forest islands". In Forest Island Dynamics in a Man-dominated Landscape. R.L. Burgess and D.M. Sharpe eds. Springer-Verlag: New York, pp.57-95. Ratti, J.T. and K.P. Reese. 1988. Preliminary test of the ecological trap hypothesis. J. Wildl. Manage. 52:484-491. 39 Rearden, J.D. 1951. Identification of waterfowl nest predators. Journal of Wildl. Manage. 15:386-395. Robinson, S.K. 1991. "Population dynamics of breeding Neotropical migrants in a fragmented Illinois landscape". In Ecology and Conservation of Neotropical Migrants. J.M. Hagan and D.W. Johnston eds. Smithsonian Institution Press: Washington, DC. pp.408-418. Robinson, S.K., F.R. Thompson, T.M. Donovan, D.R. Whitehead and J. Faabourg. 1995. Regional forest fragmentation and the nesting success of migratory birds. Science 267:1987-1990. Rosenberg, K.V. and M.G. Raphael. 1986. Effects of forest fragmentation on vertebrates in Douglas fir forests. In Wildlife 2000: Modelling Habitat Relationships of Terrestrial Vertebrates. J. Verner, M.L. Morrison, and C.J. Ralph eds. Madison: The University of Wisconsin Press, pp. 263-272. Rowe, J.S. 1972. Forest Regions of Canada. Department of the Environment. Canadian Forest Service Publication No. 1300. Rudnicky, T.C. and M.L. Hunter. 1993. Avian nest predation in clearcuts, forests, and edges in a forest-dominated landscape. J. Wildl. Manage. 57:358-364. Santos, T. and J.L. Telleria. 1992. Edge effects on nest predation in Mediterranean fragmented forests. Biological Conservation 60:1-5. SAS Institute Inc. 1990. SAS/STAT User's Guide. Version 6. Fourth Edition, Volume 2. SAS Institute Inc. Cary, NC. SAS Institute Inc. 1994. TMP Statistics and Graphics Guide Version 3. SAS Institute Inc. Cary, NC. Seitz, L.C. and D.A. Zegers. 1993. A n experimental study of nest predation in adjacent deciduous, coniferous, and successional habitats. Condor 95:297-304. Shank, C. 1986. Territory size, energetics, and breeding strategy in the Corvidae. Am. Nat. 128:642-652. Small, M.F. and M.L. Hunter. 1988 Forest fragmentation and avian nest predation in forested landscapes. Oecologia Berl. 76:62-64. Sonerud, G.A. 1985. Nest hole shift in Tengmalm's Owl Aegolius funereus as defence against nest predation involving long-term memory in the predator. J. Anim. Ecol. 54:179-192. Sonerud, G.A. and P.E. Fjeld. Long-term memory in egg predators: an experiment with a hooded crow. Ornis Scand. 18:323-325. 40 Stephens, D.W. and J.R. Krebs. 1986. Foraging Theory. Princeton University Press: Princeton NJ. Storch, I. 1991. Habitat fragmentation, nest site selection, and nest predation risk in Capercaillie. Ornis Scand. 22:213-217. Sugden, L.G., and G.W. Beyersbergen. 1986. Effect of density and concealment on American crow predation of simulated duck nests. J. Wildl. Manage. 50:9-14. SYSTAT, Inc. 1992. SYSTAT: Statistics. Version 5.2 Edition. SYSTAT Inc. Evanston, IL. Trevor, J.T., R.W. Seabloom and R.D. Sayler. 1991. Identification of mammalian predators at artificial waterfowl nests. Prairie Nat. 23:93-99. Walters, B. B. 1991. Small mammals in a subalpine old-growth forest and clearcuts. Northwest Sci. 65:27-31. Whitcomb, R.F., C.S. Robbins, J.F. Lynch, B.L. Whitcomb, M.K. Klimkiewicz and D. Bystrak. 1981. 'Effects of forest fragmentation on avifauna of the eastern deciduous forest'. In Forest Island Dynamics in Man-Dominates Landscapes. R.L. Burgess and D.M. Sharpe eds. Springer Verlag: New York, NY. pp. 125-205. Wilcove, D.S. 1985. Nest predation in forest tracts and the decline of migratory songbirds. Ecology 66:1211-1214. Wilcove, D.S., C .H. McLellan and A.P. Dobson. 1986. "Habitat fragmentation in the temperate zone". In Conservation Biology. M.E. Soule ed. Sinauer Associates: Sunderland, Massachusetts, pp. 237-256. Willebrand, T. and V. Marcstrom. 1988. On the danger of using dummy nests to study predation. Auk 105:378-379. Yahner, R.H., C.G. Mahan and C A . Delong. 1993. Dynamics of depredation on artificial ground nests in habitat managed for ruffed grouse. Wilson Bull. 105:172-179. Za rJ .H . 1984. Biostatistical Analysis. 2nd ed. Prentice-Hall Inc.: Englewood Cliffs, NJ. APPENDIX 1 Latin names of passerines nesting in the ESSF forest 41 P ine s i sk in . Carduelis pinus ( 2 2 % ) * V a r i e d thrush Ixoreus naevius ( 1 2 % ) R u b y - c r o w n e d k ing let Regulis calendula ( 12% ) H e r m i t thrush Catharus guttatus ( 9% ) Red -b reas ted nuthatch Sitta canadensis (9%) Swa inson ' s thrush Catharus ustulatus ( 8%) Y e l l o w - r u m p e d warbler Dendroica coronata (8%) D a r k - e y e d j u n c o Junco hyemalis ( 7%) G o l d e n - c r o w n e d k ing le t Regulis satrapa (5%) M o u n t a i n ch i ckadee Parus gambeli ( 4%) T o w n s e n d ' s warb ler Dendroica townsendi (2%) W i l s o n ' s warb le r Wilsonia pusilla ( 2% ) W i n t e r wren Troglodytes troglodytes ( 1%) L i n c o l n ' s Spa r row Melospiza lincolnii ( 1%) O l i v e - s i d e d f lycatcher Contopus borealis (1%) *Percent o f total number of birds heard over 5 point counts (June 5-July 9, 1993). 42 APPENDIX 2 Latin names of nest predators Deer Mouse Peromyscus maniculatus (32%)* Avian predator, possibilities include: (28%) American Crow Corvus brachyrhynchos Common Raven Corvus corax Clark's Nutcracker Nucifraga columbiana Gray Jay Perisoreus canadensis Red Squirrel . Tamiasciurus hudsonicus (17%) Yellow-pine Chipmunk Eutamias amoenus (11%) Unidentified shrew, possibilities include: (8%) Masked Shrew Sorex cinereus Dusky Shrew S. obscurus Snowshoe Hare Lepus americanus (2%) Unidentified carnivore, possibilites include: (2%) Marten Martes americana Wolverine Gulo luscus Coyote Canis latrans Red Fox Vulpes fulva *Refers to percent identified from egg remains, out of total identified. 

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.831.1-0086821/manifest

Comment

Related Items