SEARCH

SEARCH BY CITATION

Keywords:

  • faecal DNA;
  • genetic monitoring;
  • hair trapping;
  • Italian Alps;
  • non-invasive genetic sampling;
  • opportunistic sampling;
  • sampling design;
  • small populations;
  • transect sampling;
  • Ursus arctos

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

1. Non-invasive genetic sampling (NGS) of hair and faeces has become an important tool for monitoring wildlife populations, but many managers question the feasibility and cost-effectiveness of these methods for long-term monitoring. To address this question, more studies are needed that simultaneously evaluate the effectiveness and efficiency of multiple NGS designs in the same study area.

2. In 2003–2004, we carried out an experimental study of NGS for a small brown bear Ursus arctos population established by translocation in the Italian Alps. We evaluated and compared the effectiveness and efficiency of three NGS approaches including two systematic designs, baited hair traps and transect sampling of hair and faeces, and opportunistic collection of faecal and hair samples. Effectiveness was evaluated in terms of the number of samples collected, bears identified, genotyping success and error rate, detection frequencies, individual movement and spatial distribution of the species. We also evaluated the suitability of the data collected for population size estimation using single- and multi-session approaches. Efficiency was assessed by calculating total cost/genotyped sample, cost/unique bear identified and cost/bear sample.

3. During 2 years of sampling, 1164 samples and 15 unique genotypes were obtained. From these genotypes, we documented reproduction, an increase in the minimum population size of bears in the study area and important information on specific bears causing damages to property.

4. The optimal sampling strategy combined systematic hair trapping and opportunistic sampling, as the pooled data set efficiently provided large sample quantities, the highest number of identified bears, multiple individual detections, information on bear distribution and suitable data for population size estimation.

5. We provide an example of how the efficiency of NGS monitoring can be improved by integrating sampling into routine duties of existing field personnel.

6.Synthesis and applications. This study provides baseline data for monitoring brown bears in the Italian Alps and has important implications for NGS of other small populations in human-dominated landscapes. Conservation of small populations in such habitats will benefit from multiple strategies that obtain critical demographic, spatial and genetic information in a cost-effective manner.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Non-invasive genetic sampling (NGS) has become a powerful and widely used tool for studying wildlife populations of common and rare species (Waits & Paetkau 2005). Pilot studies are important for predicting sample sizes and sampling effort needed to achieve project goals. However, pilot studies for comparing the effectiveness and efficiency of multiple NGS methods have received less attention. When establishing a monitoring protocol, it is critical to know how sampling techniques perform in a particular study area, and the financial and human resources required to obtain suitable data. Assessing the relative strengths and weaknesses of different field methods can allow researchers to choose the optimal strategy and more efficiently allocate resources.

The feasibility and effectiveness of NGS depends upon many factors including the objectives of the study, sampling methods and study design, species, population density, habitat and regional climate, and ultimately the resources available. In bear research and management, where NGS is widely employed, the most common sources of DNA are hair and faeces (Taberlet et al. 1997; Woods et al. 1999). Furthermore, most studies use a single sampling technique to obtain samples. Only a few recent studies have implemented more than one technique simultaneously (Gervasi et al. 2008; Kendall et al. 2008, 2009), but no studies have evaluated accuracy and cost-effectiveness of multiple NGS methodologies in a model system where the initial population size, age and sex of individuals are known.

The recent translocation of brown bears Ursus arctos (Linnaeus 1758) in the Italian Alps offers a unique opportunity to evaluate the effectiveness of different NGS techniques for small bear populations. Between 1999 and 2002, nine bears were released into Parco Naturale Adamello Brenta (PNAB) in Trentino, northern Italy, where one to three relict bears survived. After initial radiomonitoring, NGS was selected as the preferred monitoring method for obtaining demographic, spatial and genetic information, and to assess reproductive success and population trends. Further, due to high human presence in this area, identification of the species and individuals responsible for damages was fundamental for improving the efficacy of conflict management.

An experimental project of NGS was implemented in 2003–2004 in Trentino aimed at addressing the following objectives: (i) assess the feasibility of hair and faecal DNA sampling for the small brown bear population in the Italian Alps; (ii) evaluate the effectiveness and efficiency of different field sampling designs for obtaining demographic, reproductive, genetic and spatial information, and to identify individuals responsible for damages; and (iii) establish a long-term monitoring protocol based on NGS techniques. To address these objectives, we compared three approaches: (i) systematic collection of hair samples at baited hair traps; (ii) systematic collection of faecal and hair samples on transects; and (iii) opportunistic collection of faecal and hair samples throughout the bear range. The results of this study are particularly relevant for NGS monitoring of other small populations of elusive species in human dominated environments.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Study area and population

The study area includes 1600 km2 in the western part of the Trento province (PAT) in the central Alps of northern Italy (Fig. 1). The region is characterized by a wide range of natural and human-modified habitats and elevation ranges from 100 m in some valleys bottoms to over 3500 m in the mountain ranges. The forest is dominated by deciduous trees below 1000 m but at higher elevations (1000–2000 m) conifers are dominant. Woodlands are replaced by shrubs and herbaceous plants above 2000 m. The study area is divided into numerous valleys, which are further fragmented by towns and roads. Human density is high (81 inhabitants km−2 in the PAT) especially in the valleys, where the economy is dominated by tourism and agriculture. Mid altitude areas (500–1000 m) are characterized by diffuse farming and livestock grazing. In the higher elevations (>1000 m), the human density is lower and concentrated around the main towns.

image

Figure 1.  Map of the study area and distribution of bear samples. (a) Trentino, Italy; (b) roads and towns (black) and >2000 m elevation of the Brenta range (white); (c–f) distribution of bear samples (black dots) with different techniques: (c) 2003 and (d) 2004 systematic grid and hair trap locations, (e) 2004 transects (black lines), (f) 2004 opportunistic sampling. Grey dots are hair trapping sites where bears were not detected. The grey line is PNAB boundary.

Download figure to PowerPoint

From 1999–2002, three male and six female bears were released in the eastern portion of PNAB (Fig. 1) where the last native bears survived. When this study began in 2003, all translocated bears were 4–9 years old. In addition, two female cubs were born in 2002. In the Alps, brown bears mainly use forested habitat between 300 and 1400 m and prefer deciduous and mixed forest (Mustoni 2004; Preatoni et al. 2005).

Field methods

Three sampling methods were evaluated (Table 1). Hair trapping (HT) and opportunistic sampling of hair and faeces (OP) were implemented in 2003 and 2004; transects for collection of hair and faeces (TR) only in 2004. Field work was carried out by the personnel of various agencies and by volunteers, and was organized according to standard work schedules. Hair samples were collected using sterilized forceps or latex gloves and placed in coin envelopes stored in silica desiccant or alternatively in 95% ethanol (OP only); c. 10 mL of homogenized faeces were collected using disposable spoons and stored in 40 mL of 95% ethanol. Geographic coordinates were recorded for all sample locations, except for OP in 2003. Any remaining hair or scat was removed to avoid repeated sampling.

Table 1.   Sampling methods evaluated in this study
MethodYearSampling schemeSampling unitsSampling extentSampling duration
  1. HT, hair traps; TR, transects; OP, opportunistic.

HT2003 2004Systematic, attractant Systematic, attractant39 4 × 4 km cells 41 4 × 4 km cellsc. 650 km2 grid c. 650 km2 grid8 sessions, May–Oct 7 sessions, May–Oct
TR2003Systematic17 trails145 km6 sessions, May–Oct
OP2003 2004Opportunistic Opportunistic Bear range Bear rangeYear round Year round

The study area for HT covered c. 650 km2 (Fig. 1c,d). The trapping grid was established following designs and guidelines outlined in previous DNA-based inventories in North America (Mowat & Strobeck 2000; Boulanger et al. 2002) and considering the home ranges of translocated bears. Landscape features and human presence were believed to impact bear movements and decrease the effective radius of traps. Thus, we used a 4 × 4 km grid cell size, and the grid was stratified to avoid sampling in unsuitable bear habitats (i.e. above 2000 m, towns) using ArcView 3.2 GIS (ESRI, Redlands, CA, USA), resulting in irregular shaped cells (Fig. 1; Table 1). Based on the knowledge of local bear biologists and radiotelemetry data, traps were placed in the best predicted bear habitat accessible by 4-wheel drive vehicles, a maximum of 10 min walk from vehicle and at least 200 m from major trails. Hair traps were set following the methods of Woods et al. (1999). Briefly, we placed a strand of barbed wire around several trees c. 50 cm above the ground delimiting an area of c. 25 m2, and baited the centre of the trap using 1-L mixture of rotten blood and fish poured over a central tree. In addition, we placed <100 g of corn underneath a heavy rock to increase recapture probability. In an attempt to increase detection of cubs, a second strand of barbed wire was added c. 20 cm from the ground in eight cells where females with cubs were present in 2004. Sites were visited for sample collection and for lure and corn replacement 21 days after initial setting, for eight sampling sessions in 2003 and seven in 2004 (Table 1). Midway through the sampling season, c. 50% of traps were moved at least 0·5–1 km away to increase cell coverage and minimize habituation (Boulanger et al. 2006).

Transect sampling was conducted in 2004 only (Table 1). A system of 17 transects was defined and no transects were established in the southern portion of the trapping grid (Fig. 1e) due to resource constraints and low sampling success during an exploratory study in 2003. Transect length was 4·5–17 km (mean = 8·5 km) for a total of 145 km, and each transect was surveyed in 2·5–6·5 h (mean = 3·8 h). Personnel searched for hair and faeces on trails or in the immediate surroundings. Logs or branches crossing the trail or objects with visible bear signs (trees with claw marks, destroyed ant hills, daybeds, etc.) were inspected for the presence of hair. Surveys were carried out monthly from May to October. Due to seasonal differences in the accessibility of trails, seven transects were surveyed five times and one transect was surveyed four times.

Hair and faeces were collected opportunistically throughout the bear range during all of 2003 and 2004 (Table 1, Fig. 1). Samples were collected during normal activities of agency personnel or following notification by third parties. Samples obtained from damage to property and attacks on livestock were also stored for analysis.

Genetic methods

Prior to 2003, we developed a genetic databank of translocated bears. Eighteen microsatellite loci (Taberlet et al. 1997; Paetkau, Shields & Strobeck 1998) were evaluated for eight samples from the translocated bears and their progeny and the eight loci producing the lowest probability of identity among siblings (PID(sib); Waits, Luikart & Taberlet 2001) were selected for multilocus genotyping. Genotypes were replicated a minimum of three times to serve as reliable references.

Convention on International Trade in Endangered Species of Wild Fauna and Flora (CITES) permits were obtained to import samples to the USA. In 2003 all putative bear scat and hair samples with follicles were processed for analyses. To minimize laboratory costs in 2004, scats that looked old and hair samples with <5 follicles were discarded. Extraction methods, PCR protocols and primers for microsatellite genotyping are described in Appendix S1. Sex determination was performed using the Amelogenin locus (Ennis & Gallagher 1994) as in Robinson, Waits & Martin (2007). To screen for brown bears among samples that failed microsatellite genotyping, we amplified a 150-bp segment of mitochondrial DNA (mtDNA) control region (Murphy, Waits & Kendall 2000).

We followed guidelines in Paetkau (2003) and Waits & Paetkau (2005) for ensuring microsatellite genotyping data quality and quantifying error rate; genotypes were assigned to samples using a reference genotype approach (Adams & Waits 2007; see Appendix S1 for details).

Evaluation of sampling methods

We assessed effectiveness of the sampling techniques through the following criteria: number of samples collected, bears identified, genotyping success, genotyping error rate, bear detection frequencies and spatial distribution of bear samples and individuals. We tested for significant differences in genotyping success and error rate among sampling techniques performing a logistic regression model in sas using proc genmod (sas 9.1; SAS Institute Inc., Cary, NC, USA). We used α = 0·05 and Bonferroni corrections for the set of tests of pairwise differences and contrasts. We looked at the temporal distribution of number samples collected, bears identified, new individuals detected and genotyping success to identify an optimal sampling season.

To evaluate whether hair and faecal DNA sampling methods could be used for future applications of capture–mark–recapture (CMR) approaches for this population, we analysed the 2004 data using closed population size estimators. OP data were analysed with the single-session approach of program capwire (Miller, Joyce & Waits 2005) based on a maximum likelihood estimator, implementing the likelihood ratio test to choose between the even capture probability and the two innate rates models, setting the maximum population size to 50. Systematic sampling data were analysed with multi-sessions models in program mark (White & Burnham 1999). The Huggins estimator (Huggins 1991) was used to compare models including time-varying capture probability, behavioural response and a covariate indicating whether individual bears were translocated or born in the study area to model heterogeneity in capture probability. Given the small number of animals in the population, we limited model complexity to ≤5 parameters. Model support was evaluated with the Akaike Information Criterion adjusted for small sample sizes (AICc), and population size was estimated as a derived parameter using model averaging to account for model uncertainty. We calculated log-based 95% CI for the population size estimate accounting for the minimum number of bears sampled (White, Burnham & Anderson 2001).

Efficiency of sampling methods applied in 2004 was evaluated calculating cost/genotyped sample, cost/unique bear identified and cost/bear sample, considering field (excluding transport costs) and laboratory (supplies and labour) expenses.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Genetic analyses and data quality

After microsatellite PCR screening for sample quality in 2003, we excluded 36% of hair trap samples, 85% of opportunistic hair and 73% of scats. In 2004, we culled 20% of hair trap samples, 76% of transect hair samples, 21% of opportunistic hair samples, 44% of transect scats and 51% of opportunistic scats.

The variability of the selected microsatellite loci was high with Ho = 0·80 and four to eight alleles per locus. PID(sib) at eight loci was low in both years (0·00065 in 2003, 0·0008 in 2004). The PID(sib) threshold for accepting a genotype was 0·033 in 2003 and 0·015 in 2004 (Appendix S1). As PID(sib) observed using the five loci with the lowest discriminatory ability was below 0·033 (2003) and 0·015 (2004), any sample successfully analysed with high confidence at five to eight loci was assigned a genotype.

Average per locus genotyping error rate was 3·7–7·7% for HT, 19·4% (2004 only) for opportunistic hair and 30·2–38·6% for scats. Fifteen per cent of hair samples that gave completed genotypes after one amplification were reamplified; all matched the original genotype. There were no one or two mismatch pairs in the eight-locus reference genotypes, indicating low probability of any undetected genotyping errors (Fig. S1). Multiple observations of a genotype were also taken as a measure of genotyping reliability. Individuals were detected from 1 to 59 samples in 2003 and 4 to 52 samples in 2004. The three genotypes sampled only once in 2003 were resampled in 2004.

All sex ID PCR replicates gave consistent results or matched the known sex based on field data. Forty of 181 samples that failed microsatellite genotyping in 2003 and 91 of 173 in 2004 were identified as bear samples after mtDNA fragment analysis.

Evaluation of NGS methods

In 2003, 363 samples were collected. HT provided the highest sample numbers (63%), and OP obtained similar numbers of hair and scat (Table 2a). In 2004, 801 samples were collected; HT samples were the most numerous (60%; Table 2b). TR provided equal quantities of hair and scat, whereas the number of OP faecal samples was double the number of hair samples (Table 2b).

Table 2.   Results by sampling method and sample type for (a) 2003 and (b) 2004
 Samples collectedSamples analysedUnique bears identifiedGenotyping success
HairScatTotalHairScatTotalHairScatTotalHairScatTotal
  1. HT, hair traps; TR, transects; OP, opportunistic.

(a)
HT227 227214 2148 8124 124
OP7363136716313435781119
Year total300633632856334885913211143
(b)
HT480 480229 22913 13181 181
TR353469252550233358
OP851672524313718081314304272
Year total60020180129716245913131521447261

Nine bears were identified in 2003 and 15 in 2004 (Table 2, Fig. 2). No single method detected all bears. Most bears (6 in 2003, 12 in 2004) were detected by both HT and OP, and in 2004 results from these methods combined identified all individuals that were sampled; in contrast, TR detected only two bears. With OP, more bears were identified from scat than from hair. HT and OP detected bears of all age and sex classes (Fig. 2). In 2003, one cub was sampled with HT; in 2004 a litter of two cubs was identified with opportunistic scats and a litter of three cubs with HT and opportunistic scats. The parents of all offspring were identified. All individuals sampled corresponded to translocated bears and their progeny, therefore remnant bears of the former population are assumed to be dead and no immigrants were detected. In 2004, 18 samples were collected at 10 damage sites. At eight sites, we documented the presence of a brown bear and, at six of these sites, the individual responsible for the damage was identified. Two of the bears identified were detected at >1 site.

image

Figure 2.  Detection frequency of brown bears possibly present in the study area in (a) 2003 and (b) 2004 using hair traps (HT), opportunistic hair (OPh) and scat (OPs), transect hair (TRh) and scat (TRs). Age class and sex are indicated as follows: am, adult male; af, adult female; year 1-f, 1-year-old female; year 2-f, 2-year-old female; cub m, male cub; cub f, female cub.

Download figure to PowerPoint

In 2003, HT genotyping success was higher than OP considering hair and faeces combined as well as separately (among three tests the lowest χ2 = 27·15; d.f. = 1; < 0·0001). In 2004, TR and opportunistic scats produced the lowest genotyping success rate (among eight tests the lowest χ2 = 13·71; d.f. = 1; = 0·0002); HT yielded higher success rates than OP considering hair and faeces combined (χ2 = 25·82; d.f. = 1; < 0·0001), but HT success was not significantly higher than opportunistic hair alone (Table 2b).

HT had lower error rates than faeces in 2003 (χ2 = 190·12; d.f. = 1; < 0·0001) and OP in 2004 considering hair and faeces combined as well as separately (among three tests the lowest χ2 = 45·79; d.f. = 1; < 0·0001). Among OP samples from 2004, hair produced a lower error rate than scat (χ2 = 13·75; d.f. = 1; = 0·0002). In HT samples from 2004, one mixed sample was identified by more than two alleles at multiple loci.

Detection frequency varied greatly among individuals and with different techniques (Fig. 2). Using HT, individuals were detected during one to six sessions in 2003 (mean 3·1, SD 2·2) and one to seven in 2004 (mean 3·4, SD 1·9). Capture frequency was lowest (1–2 sessions) for TR in 2004. Combining opportunistic hair and faecal detections increased detection frequencies and bears were identified one to four different times (mean 2, SD 1·2) in 2003 and one to nine times (mean 3·6, SD 2·5) in 2004.

In 2003, bear samples collected at HT were found in 38·4% of grid cells, on the eastern side of the Brenta range (Fig. 1c), and approximate locations of bear samples collected opportunistically had a similar distribution. Individuals were sampled at one to seven trap sites, and at one to four locations with OP. In 2004, bears were sampled on both sides of the Brenta range (Fig. 1d,e,f). Overall, HT and OP provided the greatest spatial coverage and were the most effective at detecting bear movement. In 2004, bear samples were found in 53·6% of the grid cells with HT and along 8 of 17 transects. The number of locations where individual bears were detected ranged from 1 to 12 for HT, 1 to 8 for OP and 2 for TR. In general, HT and OP locations of bears detected with both methods were overlapping and complementary (Fig. 3).

image

Figure 3.  Spatial distribution of multiple captures in 2004 for specific individuals: (a) adult male (am 2); (b) two adult females (af 3, af 5); (c) yearling female (1-year f) and male cub (cub m3). *, hair traps; bsl00066, transects; •, opportunistic.

Download figure to PowerPoint

For HT, the number of samples collected, individuals and new bears identified varied greatly during the sampling period (Table S1a). Fewer bears and fewer new bears were detected in the later sessions. For example, no new individuals were sampled after session IV (July) in 2003 and session V (August) in 2004. Genotyping success varied with no clear trend but was generally lowest in the last sessions.

The temporal distribution of TR was not uniform over the surveyed area, but the number of samples collected and genotyping success were low in all sessions (Table S1b). No bears were identified after August.

OP was partitioned into three main seasons of bear activity (spring, summer and autumn; Table S1c,d). Hair sampling provided the highest sample numbers in the summer. Hair genotyping success was always highly variable, but generally was higher in the spring. More individuals were identified from hair during spring and no new bears were identified after August. Scat samples were more numerous in spring and autumn; scat genotyping success varied between years but was slightly higher in autumn than during other seasons. The number of individuals and new individuals identified was the lowest in the summer.

TR was not included in the population estimation analysis due to low recaptures. To maximize closure assumptions, we considered the first five HT sessions, and the sampling period May–August for OP (minimum count dropped to 9). The likelihood ratio test in capwire indicated individual heterogeneity (= 0·002) in OP data, and the two innate rates model was used. capwire population size estimate for 2004 OP data was 13 (95% CI: 9–18). Supported models in mark (AICc ≤2) (Burnham & Anderson 2002) for HT did not indicate time variation in capture probability or behavioural response, and included the effect on capture probability of whether a bear was translocated or born in the area (Table 3). The estimate of capture probability from the top-ranked model was 0·54 (SE 0·069) and the model-averaged population size estimate was 14 (95% CI: 13–18).

Table 3.   Model selection results from Huggins closed capture models in mark for one to five hair trapping sessions in 2004
ModelAICcΔAICcAICc weightModel likelihoodNo. parametersDeviance
  1. p, capture probability; c, recapture probability for behavioural response models; trans, an effect due to whether animals were translocated or born in the study area; (.), constant p or c; t, time varying p.

p(trans)89·3500·000·460311·0000285·156
p(.)90·9251·580·209420·4550188·861
p(.), c(.)91·9402·590·126030·2738287·747
p(trans), c(.)92·0702·720·118110·2566385·677
p(trans), c(trans)92·8223·470·081120·1762484·155
p(t)98·3929·040·005010·0109587·375

In 2004, c. €71,000 was spent to implement hair and faecal sampling and to perform genetic analyses (Table 4). HT required the most resources in the field and in the laboratory. Field costs were higher than laboratory costs, except for OP. OP provided the lowest cost/genotyped sample (€125), cost/unique bear (€645) and cost/bear sample (€64), whereas TR costs were the highest. Within OP, hair had lower cost/genotype and cost/unique bear than scat, whereas cost/bear sample was slightly higher for hair than scat. The opposite trend of hair and scat costs was observed for TR.

Table 4.   Costs (€) per sampling method estimated in 2004
 FieldLabTotalCost/genotypeCost/unique bearCost/bear sample
HairScatTotalHairScatTotalHairScatTotal
  1. HT, hair traps; TR, transects; OP, opportunistic.

HT31 417780339 220217 2173017 3017192 192
TR18 635418022 8156313436528529470727576059971212617
OP26076425903280157125298506645706164

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Evaluation of NGS methods

The current alpine landscape of southern Europe is highly modified by human presence, and a long history of human persecution has affected the behaviour of the European brown bear (Swenson et al. 2000). This and the small size of the studied population raised concerns about the feasibility of hair and faecal DNA sampling in this region and made it necessary to adapt the sampling design. Our results demonstrate that an NGS approach is feasible for the brown bear in the Italian Alps, as 1164 samples were collected over 2 years and 15 genotypes were identified corresponding to seven of the translocated bears (two males, five females) and eight of their progeny born between 2002 and 2004 (three males, five females). Our results also illustrate that effectiveness and efficiency is maximized by combining the systematic design of HT and OP (Table 5).

Table 5.   Strengths and weaknesses of non-invasive genetic sampling methods for the bear population ranked from 1 to 3 (1 = highest rank)
CriterionHTTROP
  1. HT, hair traps; TR, transects; OP, opportunistic.

Sample quantity132
Sample quality132
Bears detected131
Detection frequency131
Spatial inference132
Population size estimation131
Cost/genotype231
Cost/unique bear231
Cost/bear sample231
Mean score1·331·3

HT and OP provided large sample quantities, detected the highest number of bears, allowed multiple detections of individuals, and showed an extensive overlap of individuals and spatial distribution (Table 5). In addition, the HT and OP population size point estimates differed by only one bear, were close to the minimum count from both methods combined, and 95% CI were small. Further, similar to the results of Kendall et al. (2009), both methods detected all age and sex classes, allowing adequate population representation for population size estimation. The good agreement between the information from these methods suggested that most or all bears in the study area were sampled in 2003 and 2004. Combined information from HT and OP was especially important for increasing detections of females and cubs and providing detailed data on individual movements. Our results confirmed that the population increased through reproduction of translocated animals, and that translocated bears settled in a small area previously occupied by the last indigenous bears. Thus, our study, like Kasworm, Proctor & Servheen (2007), demonstrates the value of NGS for monitoring success of the early stages of translocation and population augmentation.

HT was logistically challenging and resulted in high costs, but offered several benefits compared to the other methods (Table 5). Sampling success and sample DNA quality were the highest due to the collection of plucked hair at regular intervals, in contrast to shed hair and scat (Murphy et al. 2000; Morin et al. 2001) mostly found with OP and TR. Capture probability was higher than other bear hair trapping projects in North America perhaps due to a smaller sampling grid (c. 16 km2 compared to ≥25 km2) (i.e. Boulanger et al. 2002; Kendall et al. 2009) that resulted in a higher encounter rate, and is suitable for population estimation with this method (i.e.  0·2) (White et al. 1982; Boulanger et al. 2008). We used a small amount of corn as a nutritional reward to maximize bear detections. Although we did not detect a behavioural response in capture probability, the use of a reward could introduce a bias and is generally avoided for CMR estimation.

The greatest advantage of OP was identifying a similar number of bears and high recaptures at lower cost and effort than HT (Table 5). OP sampling success was probably due to the large overlap of the bear and human population and high management effort in the study area that facilitated collection year round. OP may be less efficient and effective in less managed wilderness areas. Combining data from hair and scat increased the spatial sampling coverage, and bear detections. This was important for achieving suitable sample sizes for capwire when using only 4 months of OP data, as in small populations (N ≤ 25) an average of >2·5 observations per individual is necessary for accurate estimates (Miller et al. 2005).

An important goal of OP was confirming bear damage and providing the documentation needed for compensation. NGS has been implemented to identify canids causing damage to livestock and to minimize conflicts with humans in other regions (Valière et al. 2003; Blejwas et al. 2006). In this study, identification to the individual level demonstrated that only a few bears caused multiple damages and allowed management actions targeting individuals.

TR was not an efficient method compared to HT and OP (Table 5). However, scat encounter rate (0·23 samples km−1) was higher than a sign survey study for the small, low-density brown bear population in Spain (Clevenger & Purroy 1996) and was within the range observed for a larger, high-density brown bear population in Glacier National Park (Kendall et al. 1992), suggesting that TR can be a valuable source of bear samples. The main weakness was DNA degradation over the month-long survey period that prevented individual ID due to low genotyping success rates (Murphy et al. 2007; Santini et al. 2007).

Unequal sampling effort most likely affected the results of sample number and bear detections, particularly for comparing HT and TR. However, we believe this did not affect the conclusions on the relative cost-effectiveness of these methods. TR data suggest that, to increase sample sizes and DNA quality for individual ID with this method, effort greater than that applied for HT would be necessary, due to the lower-quality DNA from shed hair and faeces mostly found on TR. Another challenge in comparing the effectiveness and efficiency of different sampling methods is that the area covered by each method was not entirely overlapping. However, this had minimal impact on our results because the area where all methods were applied was included in the home ranges of translocated bears in 1999–2003.

NGS for monitoring the brown bear population in the Alps

Obtaining demographic, spatial and genetic information, and assessing reproductive success and population trends are critical for management and conservation of the small brown bear population in northern Italy (IUCN 1995; Swenson et al. 2000). In the modern human-modified alpine habitat, identification of problem bears for mitigation of human-bear conflicts is also a priority. Based on the results of our 2-year pilot study, HT and OP could provide all of this critical information, and their combined application is the most promising and cost-effective strategy for monitoring this population.

The joint application of systematic and opportunistic designs offers many advantages and avoids potential bias introduced by the use of a single method. For example, systematic HT allows probability of occurrence modelling and inference of population trends and variables affecting spatial distribution (Apps et al. 2004; Boulanger, Himmer & Swan 2004) that could not be obtained with opportunistic designs alone due to unequal sampling intensity or preferential sampling at some locations. In contrast, OP is important for conflict resolution and can detect bear presence and movements in areas not covered by HT due to constraints in human-dominated environments. Further, Boulanger et al. (2008) discussed the benefits of combining multiple data sources to improve CMR population estimates, and recent studies provide empirical examples of how data from different techniques can be analysed in such a framework (Gervasi et al. 2008; Kendall et al. 2008, 2009).

One important aspect of this project was quantification of the costs of implementing an NGS-monitoring programme for this population. Although costs are substantial, the comprehensive information gathered by hair and faecal DNA sampling cannot be obtained more cheaply with traditional methods based on trapping and radiotelemetry. For example, €35 410 was spent radiomonitoring two bears in our study area during 2008 using GPS collars. Although this provided critical fine-scale information and close monitoring of two problem bears, NGS represents a more cost-effective option for gaining both population and individual data. Our data suggest that HT cost/genotype and cost/bear sample could be decreased by sampling from the end of May to mid-August, when capture probability is maximized, and avoiding the second lower strand of barbed wire that was not effective at sampling new bears, consistent with results of Boulanger et al. (2006). Further, OP laboratory costs can also be minimized by analysing only higher DNA quality samples, such as hair collected in the spring and scat collected in the autumn.

Based on our results, we have established a yearly NGS programme for the short term (4 years) that includes simultaneous implementation of OP and HT. In the long term, we propose a less-intensive monitoring regime involving yearly OP to track reproduction, immigration, genetic diversity, minimum count and population size estimation as well as periodic HT (i.e. 2–10 year intervals) combined with OP for population size estimation, spatial and trend inferences.

Implications for monitoring small populations

The utility of DNA-based wildlife monitoring is increasingly recognized (Schwartz, Luikart & Waples 2007). Long-term monitoring requires that programmes can be implemented easily and relatively inexpensively, and techniques minimizing the disturbance of endangered animals are preferred. Our findings support the application of NGS as a cost-effective and powerful tool to monitor other small or translocated populations of elusive species in human-dominated environments. Wildlife monitoring in such habitats can benefit from the integration of different designs for data collection that take into consideration the ecology and behaviour of the animals in the local environment.

In intensively managed areas, NGS can be implemented as part of the regular activities of field personnel to minimize labour costs. We recommend extensive pre-season training of the personnel conducting the sampling, including information about the objectives of the study, and regular updates on results. We stress the importance of conducting pilot studies to determine which sampling design and method or combination of methods allows efficient allocation of resources while also achieving project objectives. In this context, an adaptive management approach that allows flexibility is recommended (Nichols & Williams 2006; Schwartz et al. 2007). Ultimately, small populations can benefit from comprehensive monitoring programmes that integrate data from different standardized approaches such as NGS and radiotelemetry, and less formalized data sources such as observational sightings and bear signs (Apps et al. 2004).

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Marta De Barba received financial support from the DeVlieg Foundation, the International Bear Association and PAT. Housing and work space in the field was provided by PNAB. Laboratory costs were covered by PAT. Invaluable field assistance was provided by the Brown Bear Research and Conservation Group and rangers of PNAB; rangers of the Trento Hunting Association; C. Groff, A. Caliari and volunteers E. Guella, M. Vettorazzi and R. Flatz, and volunteers provided lab assistance. Thanks to M. Proctor, J. Boulanger, K. Kendall and J. Stetz for discussions on sampling design and population monitoring, and C. Williams for statistical advice. We also thank two anonymous reviewers and the Waits lab group for comments and suggestions for manuscript improvement.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Adams, J.R. & Waits, L.P. (2007) An efficient method for screening faecal DNA genotypes and detecting new individuals and hybrids in the red wolf (Canis rufus) experimental population area. Conservation Genetics, 8, 123131.
  • Apps, C., McLellan, B., Woods, J.G. & Proctor, M.F. (2004) Estimating grizzly bear distribution and abundance relative to habitat and human influence. Journal of Wildlife Management, 68, 138152.
  • Blejwas, K.M., Williams, C.L., Shin, G.T., McCullough, D.R. & Jaeger, M.M. (2006) Salivary DBA evidence convicts breeding male coyotes of killing sheep. Journal of Wildlife Management, 70, 10871093.
  • Boulanger, J., White, G.C., McLellan, B.N., Woods, J., Proctor, M. & Himmer, S. (2002) A Meta-analysis of grizzly bear DNA mark-recapture projects in British Columbia, Canada. Ursus, 13, 137152.
  • Boulanger, J., Himmer, S. & Swan, C. (2004) Monitoring of grizzly bear population trends and demography using DNA mark-recapture methods in the Owikeno Lake area of British Columbia. Canadian Journal of Zoology, 82, 12671277.
  • Boulanger, J., Proctor, M., Himmer, S., Stenhouse, G., Paetkau, D. & Cranston, J. (2006) An empirical test of DNA mark-recapture sampling strategies for grizzly bear. Ursus, 17, 149158.
  • Boulanger, J., Kendall, K.C., Stetz, J., Roon, D., Waits, L.P. & Paetkau, D. (2008) Multiple data sources improve DNA-based mark-recapture population estimates of grizzly bears. Ecological Applications, 18, 577589.
  • Burnham, K.P. & Anderson, D.R. (2002) Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach. Springer-Verlag, New York, NY, USA.
  • Clevenger, A.P. & Purroy, F.J. (1996) Sign surveys for estimating trends of a remnant brown bear Ursus arctos population in northern Spain. Wildlife Biology, 2, 275281.
  • Ennis, S. & Gallagher, T.F. (1994) A PCR-based sex-determination assay in cattle based on the bovine amelogenin locus. Animal Genetics, 25, 425427.
  • Gervasi, V., Ciucci, P., Boulanger, J., Posillico, M., Sulli, C., Focardi, S., Randi, E. & Boitani, L. (2008) A preliminary estimate of the Apennine brown bear population size based on hair-snag sampling and multiple data source mark-recapture Huggings models. Ursus, 19, 105121.
  • Huggins, R.M. (1991) Some practical aspects of a conditional likelihood approach to capture experiments. Biometrics, 47, 725732.
  • IUCN (1995) Guidelines for Re-introductions. IUCN/SSC Reintroduction Specialist Group, Gland, Switzerland.
  • Kasworm, W.F., Proctor, M.F. & Servheen, C. (2007) Success of grizzly bear population augmentation in Northwest Montana. Journal of Wildlife Management, 71, 12611266.
  • Kendall, K.C., Metzgar, L.H., Patterson, D.A. & Steele, B.M. (1992) Power of sign surveys to monitor population trends. Ecological Applications, 2, 422430.
  • Kendall, K.C., Stetz, J.B., Roon, D.A., Waits, L.P., Boulanger, J.B. & Paetkau, D. (2008) Grizzly bear density in Glacier National Park, Montana. Journal of Wildlife Management, 72, 16931705.
  • Kendall, K.C., Stetz, J., Boulanger, J., Macleod, A.C., Paetkau, D. & White, G. (2009) Demography and genetic structure of a recovering grizzly bear population. Journal of Wildlife Management, 73, 317.
  • Miller, C.R., Joyce, P. & Waits, L.P. (2005) A new method for estimating the size of small populations from genetic mark-recapture data. Molecular Ecology, 14, 19912005.
  • Morin, P.A., Chambers, K.E., Boesch, C. & Vigilant, L. (2001) Quantitative polymerase chain reaction analysis of DNA from noninvasive samples for accurate microsatellite genotyping of wild chimpanzees (Pan troglodytes verus). Molecular Ecology, 10, 18351844.
  • Mowat, G. & Strobeck, C. (2000) Estimating population size of grizzly bears using hair capture, DNA profiling, and mark-recapture analysis. Journal of Wildlife Management, 64, 183193.
  • Murphy, M.A., Waits, L.P. & Kendall, K.C. (2000) Quantitative evaluation of fecal drying methods for brown bear DNA analysis. Wildlife Society Bulletin, 28, 951957.
  • Murphy, M.A., Kendall, K.C., Robinson, A. & Waits, L.P. (2007) The impact of time and field conditions on brown bear (Ursus arctos) faecal DNA amplification. Conservation Genetics, 8, 12191224.
  • Mustoni, A.. (2004) L’orso bruno sulle Alpi – Biologia, comportamento e rapporti con l’uomo. Nitida Immagine Editrice, Cles (TN).
  • Nichols, J.D. & Williams, B.K. (2006) Monitoring for Conservation. Trends in Ecology and Evolution, 21, 668673.
  • Paetkau, D. (2003) An empirical exploration of data quality in DNA-based population inventories. Molecular Ecology, 12, 13751387.
  • Paetkau, D., Shields, G.F. & Strobeck, C. (1998) Gene flow between insular, coastal and interior populations of brown bears in Alaska. Molecular Ecology, 7, 12831292.
  • Preatoni, D., Mustoni, A., Martinoli, A., Carlini, E., Chiarenzi, B., Chiozzini, S., Van Dongen, S., Wauters, L.A. & Tosi, G. (2005) Conservation of brown bear in the Alps: space use and settlement behavior of reintroduced bears. Acta Oecologica, 28, 189197.
  • Robinson, S., Waits, L.P. & Martin, I.D. (2007) Evaluating population structure of black bears on the Kenai Peninsula using mitochondrial and nuclear DNA analyses. Journal of Mammology, 88, 12881299.
  • Santini, A., Lucchini, V., Fabbri, E. & Randi, E. (2007) Ageing and environmental factors affect PCR success in wolf (Canis lupus) excremental DNA samples. Molecular Ecology Notes, 7, 955961.
  • Schwartz, M.K., Luikart, G. & Waples, R.S. (2007) Genetic monitoring as a promising tool for conservation and management. Trends in Ecology and Evolution, 22, 2533.
  • Swenson, J.E., Gerstl, N., Dahle, B. & Zedrosser, A. (2000) Action plan for the conservation of the Brown Bear (Ursus arctos) in Europe. Council of Europe, Strasbourg, France.
  • Taberlet, P., Camarra, J.J., Griffin, S., Uhres, E., Hanotte, O., Waits, L.P., Paganon, C., Burke, T. & Bouvet, J. (1997) Non-invasive genetic tracking of the endangered Pyrenean brown bear population. Molecular Ecology, 6, 869876.
  • Valière, N., Fumagalli, L., Gielly, L., Miquel, C., Lequette, B., Poulle, M., Weber, J., Arlettaz, R. & Taberlet, P. (2003) Long-distance wolf recolonization of France and Switzerland inferred from non-invasive genetic sampling over a period of 10 years. Animal Conservation, 6, 8392.
  • Waits, L.P. & Paetkau, D. (2005) Noninvasive genetic sampling tools for wildlife biologists: a review of applications and recommendations for accurate data collection. Journal of Wildlife Management, 69, 14191433.
  • Waits, L.P., Luikart, G. & Taberlet, P. (2001) Estimating the probability of identity among genotypes in natural populations: cautions and guidelines. Molecular Ecology, 10, 249256.
  • White, G.C. & Burnham, K.P. (1999) Program MARK: survival estimation from populations of marked animals. Bird Study (Supplement), 46, 120138.
  • White, G.C., Anderson, D.R., Burnham, K.P. & Otis, D.L. (1982) Capture-Recapture and Removal Methods for Sampling Closed Populations. La-8787-NERP, Los Alamos National Laboratory, Los Alamos, NM.
  • White, G.C., Burnham, K.P. & Anderson, D.R. (2001) Advanced features of program MARK. Integrating People and Wildlife for a Sustainable Future: Proceedings of the Second International Wildlife Management Congress (eds R.Fields, R.J.Warren, H.Okarma & P.R.Seivert), Vol. 7.5, pp. 368377. Wildlife Society, Bethesda, MD, USA, Gödöllö, Hungary.
  • Woods, J.G., Paetkau, D., Lewis, D., McLellan, B.N., Proctor, M. & Strobeck, C. (1999) Genetic tagging of free-ranging black and brown bears. Wildlife Society Bulletin, 27, 616627.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Appendix S1. Genetic methods.

Fig. S1. Mismatch probability distribution.

Table S1. Temporal distribution of sampling methods

As a service to our authors and readers, this journal provides supporting information supplied by the authors. Such materials may be re-organized for online delivery, but are not copy-edited or typeset. Technical support issues arising from supporting information (other than missing files) should be addressed to the authors.

FilenameFormatSizeDescription
JPE_1752_sm_AppendixS1.doc94KSupporting info item
JPE_1752_sm_FS1.doc62KSupporting info item
JPE_1752_sm_TS1.doc82KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.