SEARCH

SEARCH BY CITATION

Keywords:

  • abandonment;
  • bird;
  • corticosterone;
  • Falco sparverius ;
  • land use;
  • nest success;
  • roads;
  • urbanization

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  1. The rapid increase of human activity in wild and developed areas presents novel challenges for wildlife. Some species may use human-dominated landscapes because of favourable resources (e.g. high prey availability along roadsides); however, use of these areas may increase exposure to anthropogenic stressors, such as human disturbance or noise, which can negatively affect reproduction or survival. In this case, human-dominated landscapes may act as an ecological trap.
  2. We evaluated whether American kestrel Falco sparverius reproductive failure was associated with human disturbance (traffic conditions and land development) or other common predictors of reproductive outcome, such as habitat and clutch initiation date. Also, we examined relationships among human disturbance, corticosterone (CORT) concentrations and nest abandonment to explore potential mechanisms for stress-induced reproductive failure.
  3. Twenty-six (36%) of 73 kestrel nesting attempts failed and 88% of failures occurred during incubation. Kestrels nesting in higher disturbance areas were 9·9 times more likely to fail than kestrels nesting in lower disturbance areas. Habitat and clutch initiation date did not explain reproductive outcome.
  4. Females in higher disturbance areas had higher CORT and were more likely to abandon nests than females in lower disturbance areas. There was no relationship between male CORT and disturbance or abandonment. Females spent more time incubating than males and may have had more exposure to anthropogenic stressors. Specifically, traffic noise may affect a cavity-nesting bird's perception of the outside environment by masking auditory cues. In response, incubating birds may perceive a greater predation risk, increase vigilance behaviour, decrease parental care, or both.
  5. Synthesis and applications. Proximity to large, busy roads and developed areas negatively affected kestrel reproduction by causing increased stress hormones that promoted nest abandonment. These results demonstrate that species presence in a human-dominated landscape does not necessarily indicate a tolerance for anthropogenic stressors. Managers should carefully consider or discourage projects that juxtapose favourable habitat conditions with areas of high human activity to decrease risk of ecological traps. Noise mitigation, while locally effective, may not protect widespread populations from the pervasive threat of traffic noise. Innovative engineering that decreases anthropogenic noise at its source is necessary.

Introduction

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

Human activities in developed and wild areas have led to increased disturbance in terrestrial (Papouchis, Singer & Sloan 2001; Schlesinger, Manley & Holyoak 2008) and marine (Weilgart 2007) environments. Humans and products of human activity (i.e. noise) may elicit behavioural or physiological responses because organisms perceive humans as potential predators (Beale & Monaghan 2004) or, more broadly, they may perceive human-dominated landscapes as having a higher predation risk (Frid & Dill 2002; Quinn et al. 2006). Individuals may cope with disruptive stimuli by shifting energy and time away from behaviours such as foraging (Gill, Sutherland & Watkinson 1996; Williams, Lusseau & Hammond 2006) or parental care (Fernández & Azkona 1993; Baudains & Lloyd 2007). Over time, changes in individual behaviour and physiology may lead to changes in population distribution (Rodríguez-Prieto & Fernández-Juricic 2005) or abundance (French et al. 2011). Organisms may be susceptible to disturbance because human activity may be difficult to predict (Webber, Heath & Fischer 2013) or formerly reliable cues that animals use to assess habitat quality may not be indicative of human disturbance (i.e. an ecological trap; Schlaepfer, Runge & Sherman 2002). Management of human disturbance requires a better understanding of the mechanisms by which disturbance affects reproduction and survival (French et al. 2011) and an evaluation of disturbance effects relative to other (natural or anthropogenic) factors that contribute to population change.

Human disturbance may act as a stressor and stimulate the hypothalamic–pituitary–adrenal axis, resulting in increased glucocorticoid concentrations. For example, wolves Canis lupus (Creel et al. 2002) and capercaillie Tetrao urogallus (Thiel et al. 2008) exposed to human disturbance had elevated faecal glucocorticoids. Glucocorticoids may facilitate changes to cope with stressful conditions (e.g. escape behaviour, increased foraging, gluconeogenesis) while suppressing other activities, such as reproduction (Harvey et al. 1984; Wingfield et al. 1998). In birds, elevated corticosterone (CORT, the avian glucocorticoid) may reduce parental care (Miller, Vleck & Otis 2009) or promote nest abandonment (Silverin 1998). However, there is limited evidence linking human disturbance, glucocorticoids and fitness consequences within a single study system (Cyr & Romero 2007; Busch & Hayward 2009). To our knowledge, only one study has showed an association among tourism, increased glucocorticoids and decreased reproductive performance (Ellenberg et al. 2007). Direct evidence of links between disturbance, stress and reproduction or survival may be difficult to detect because susceptibility to human disturbance may vary across species, individuals or stages of the annual cycle (i.e. breeding vs. non-breeding; Busch & Hayward 2009).

American kestrels Falco sparverius (Linnaeus 1758) are a widespread, cavity-nesting falcon that occurs in a variety of human-dominated landscapes including urban, suburban and agricultural habitats (Smallwood & Bird 2002). Kestrels are considered a human-tolerant species (Smallwood & Bird 2002); however, use of human-dominated areas may make kestrels susceptible to noxious anthropogenic stimuli that negatively affect survival or reproductive success. Further, kestrel populations are declining in parts of their range, and the cause of this decline is poorly understood (Farmer & Smith 2009). Our objective was to explore the hypothesis that human disturbance affects American kestrel reproduction and to compare human disturbance effects with other possible predictors of kestrel reproductive outcome, such as habitat characteristics (Rohrbaugh & Yahner 1997) and clutch initiation date (K. Steenhof and J. A. Heath, unpublished data). We used an information theoretical framework to evaluate whether human disturbance (land development and traffic conditions), the proportion of native shrub-steppe habitat within a kestrel territory, timing of clutch initiation or a combination of these factors predicted reproductive failure. In addition, we hypothesized that a relationship between human disturbance and reproductive failure would be associated with changes in baseline CORT concentrations. We predicted that kestrels nesting in areas of high human disturbance would have increased CORT, high rates of nest abandonment and, consequently, high rates of reproductive failure.

Materials and methods

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

American Kestrel Reproduction

American kestrel reproduction in south-west Idaho (Fig. 1) has been monitored for over 20 years as part of a long-term demography and dispersal study (Steenhof & Peterson 2009). We conducted this investigation in the breeding seasons of 2008 and 2009. We monitored 89 nest boxes posted 2·5–10 m from the ground on road signs along an expressway (Interstate 84, n = 18) or on posts and trees along secondary roads in suburban (n = 10), rural-residential (n = 24), agricultural, (n = 22) and shrubland (n = 15) areas. Beginning in early March, we visited boxes every 7–10 days to determine occupancy and clutch initiation date (the date the first egg was laid). Kestrels typically laid one egg every other day until the clutch was complete with 4–6 eggs. If we discovered a nest with more than one egg, or a nest with a complete clutch (4–6 eggs and on subsequent visits there were no additional eggs), we backdated to estimate clutch initiation date.

image

Figure 1. Map of the continental United States of America with Idaho outlined in bold. The inset shows the area in south-west Idaho where we studied the effects of traffic conditions and human development on American kestrel reproductive outcome. The black dots represent occupied nest boxes, light grey lines are secondary roads, and the dark grey line is an expressway (Interstate 84). Boise is given as a reference point.

Download figure to PowerPoint

Once the clutch was complete, we revisited nest boxes as often as necessary (every 1–2 days) to capture adults during incubation. We captured adult kestrels by plugging the nest box entrance and removing birds from the box with our hands. We collected approximately 0·4 mL of blood from the jugular vein with a 26½-gauge needle and syringe and stored samples in heparinized vials placed on ice. We collected blood samples within 5 min. A handling time of < 3 min has been suggested for obtaining baseline CORT (Romero & Reed 2005); however, there is not a significant increase in CORT for kestrels bled in < 5 min (Strasser & Heath 2011). We collected blood samples between 0900 and 1400 to minimize time-of-day effects on CORT (Wingfield, Vleck & Moore 1992). We marked kestrels with United States Geological Survey aluminium bands, collected measurements and placed the bird back in the nest box in < 15 min. In the laboratory, blood samples were centrifuged at 17 000 g for 15 mins to separate plasma from the cellular fraction. Plasma was stored at −80 °C until analysis by enzyme-linked immunosorbent assays (ELISA; Cayman Chemicals, Ann Arbor, Michigan, USA) for CORT as described in Strasser & Heath (2011). Interassay CORT variation averaged 8·6%, and average intra-assay variation was 2·0%.

After capturing adults, we returned to nest boxes on estimated hatch dates (5–15 days after adult capture). Nests were checked when the oldest nestling was 10 days old and again at 25 days. We considered kestrel pairs successful if they produced at least one 25-day-old nestling (85% of fledging age). If we noted reproductive failure, we determined cause based on the presence or absence of kestrel eggs, nestlings or adult birds. We assigned the cause of failure as predation if a nest that previously contained a full clutch had no signs of eggs or nestlings. We assumed that depredation occurred while adults continued to attend the nest. A nest was categorized as abandoned if it contained intact, cold eggs and no adults were present during two subsequent nest checks performed over 2–3 days. One day after a nest appeared abandoned, we conducted roadside surveys by walking 900-m transects parallel to the road in either direction from the nest box on both road sides to determine whether adults collided with vehicles. We accounted for scavenger bias and carcass persistence in our estimate of road mortality using marked quail Coturnix coturnix (Linnaeus 1758) placed along roadsides (Antworth, Pike & Stevens 2005). Quail were used because they were similar in size to kestrels and carcasses were readily available. We had a high probability of carcass detection (0·75) and carcasses persisted 14–49 days (Strasser 2010).

Human Disturbance and Habitat Characteristics

We created a 900-m radius buffer (typical kestrel territory size, K. Steenhof, unpublished data) centred on GPS points for each occupied nest box within a geographic information system (ArcGIS 9.2, ESRI 2009). We quantified the proportion of human development and three different vegetation communities: shrub-steppe, agriculture and introduced vegetation within the 900-m buffer using satellite imagery of south-west Idaho (http://www.gap.uidaho.edu).

Human disturbance at each nest was estimated from the proportion of human development within a territory and traffic conditions on nearby roads (Frid & Dill 2002; Parris & Schneider 2008). We recorded the number of lanes, posted roadway speed (kph) and average number of vehicles per day (http://www.compassidaho.org, http://itd.idaho.gov) on the road nearest to each occupied nest box. We used a principal component analysis based on these four variables to create a disturbance index that combined multiple, correlated variables into uncorrelated principal components (PC) that account for variation within the data (Manly 1994). Each nest received a disturbance PC score based on its proportion of developed land and traffic conditions. Nests in areas with larger, busy roads and developed landscapes received a high disturbance PC score, and nests in areas with smaller, less busy roads in undeveloped landscapes received a lower disturbance PC score (see 'Results'). We estimated the distance between each nest box to the edge of the nearest road using a laser rangefinder (Bushnell Yardage Pro). Distance to road was grouped into three ordinal categories on the basis of natural breaks in the distance from road distribution. We scored nest boxes ≤ 17·5 m from the road as ‘1’, nest boxes > 17·5 m and < 34 m as ‘2’ and nest boxes ≥ 34 m from the road as ‘3’.

Statistical Analysis

We used generalized linear models with a logit-link function to predict reproductive outcome (success or failure). The predictor variables for the disturbance hypothesis were disturbance PC score and distance between the nest box and the road. Proportions of shrub-steppe, agriculture and introduced vegetation were correlated; thus, we used only the proportion of shrub-steppe as a predictor variable to represent habitat characteristics. We used a standardized clutch initiation date to represent the timing hypothesis because of annual variation in nesting phenology (Steenhof & Heath 2009). The standardized date was calculated for each nest by dividing its clutch initiation date by the median clutch initiation date for all nests in that year. We used Akaike Information Criterion adjusted for small sample size (AICc) to compare evidence for all possible combinations of our research hypotheses (Burnham & Anderson 2002). We considered models with delta-AIC < 2·0 as good candidate models and calculated model-averaged parameter estimates and unconditional standard errors (Burnham & Anderson 2002). We calculated 85% confidence intervals for each parameter estimate to be consistent with the information theoretical framework of the analysis (Arnold 2010). We examined residuals to ensure that data met model assumptions. We did not include five nests in our reproductive outcome analysis. One was a nest in a low disturbance area that was abandoned after a nearby construction project, thus a disturbance PC score based on traffic did not adequately capture the surrounding environment. Another excluded nest failed because a storm destroyed the box. Also, we did not use three nests where an adult died from vehicle collision because these data would overestimate the disturbance effect. We evaluated relationships among disturbance PC scores, habitat characteristics and standardized clutch initiation date using Spearman's correlations.

We used a generalized linear model to evaluate whether the disturbance PC score explained baseline CORT. We analysed male and female CORT separately (Bonier et al. 2007). We checked whether capture date or time of day influenced baseline CORT using linear regressions. We used a generalized linear model with a logit-link function to examine whether female or male CORT predicted kestrel reproductive outcome and generated parameter estimates and 95% confidence intervals to be consistent with a P-value evaluation. For this analysis, we did not include nests that were depredated because we were interested in the role of adult CORT as a mechanism explaining abandonment. All analyses were conducted in SAS 9.2 (SAS Institute 2002). Descriptive statistics are presented as mean ± SD unless otherwise noted.

Results

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

American Kestrel Reproduction

American kestrels made 36 and 37 nest attempts in 2008 and 2009, respectively (Table 1). The return rate for adults was low (7%), and no pairs nested in the same nest box in both years. Clutch initiation dates ranged from 29 March–3 June in 2008 and 19 March–11 July in 2009. Completed clutch sizes were lower in 2008 (4·8 ± 0·8 eggs) than 2009 (5·2 ± 0·6 eggs, Wilcoxon Z = −2·09, P = 0·04); however, the number of eggs that successfully hatched did not differ between years (4·3 ± 1·2 hatchlings per brood). Also, productivity (3·9 ± 1·3 fledglings per successful pair) did not differ between years. Reproductive success averaged 64%, with 72% (2008) and 57% (2009) of nesting attempts producing at least one fledgling. Most (n = 23, 88%) of the nests that failed did so during incubation. Only three nests failed during the nestling stage. Sixteen of the 26 failed nests (62%) were abandoned.

Table 1. Number of nest attempts, reproductive outcome and causes of failure for American kestrels in south-western Idaho, 2008–2009. Causes of failure included abandonment, road mortality and other (probable cases of predation, hatching failure, weather and construction)
YearNest attemptsReproductive outcomeFailed egg stageFailed nestling stage
Success (%)Failure (%)AbandonedRoad mortalityOtherRoad mortalityOther
  1. a

    Nests that failed as a result of adult road mortality, a nest destroyed by weather, as well as a nest where adults abandoned and renested after nearby construction were not included in analyses of nest outcome (n = 5).

20083626 (72)10 (28)80011
20093721 (57)16 (43)82501
Totala7347 (64)26 (36)162512

Human Disturbance, Habitat Characteristics and Clutch Initiation Date Effects on Reproductive Outcome

The disturbance PC accounted for 80% of the total variation in traffic and developed land values from occupied nest boxes and was positively associated with high numbers of vehicles, speed, number of lanes and proportion of human development (Table 1 in Strasser & Heath 2011). Disturbance PC scores for nest boxes ranged from −2·40 to 2·57 with a natural break near −0·78. Higher PC scores (> −0·78) represented, on average, more vehicles (14 192 per day ± 1593), faster vehicles (109 kph ± 4·5), more lanes (3·2 ± 0·2) and more human development (27% ± 4·3) than low disturbance boxes, which had fewer (1221 per day ± 191) and slower vehicles (68 kph ± 2·1), fewer lanes (1·9 ± 0·1) and less human development (5% ± 0·3).

There was no correlation between disturbance PC score and percentage of agriculture (rs = −0·17, P = 0·15) or sagebrush steppe (rs = −0·12, P = 0·30) within a territory. Territories with lower proportions of introduced grassland had significantly higher disturbance PC scores (rs = −0·35, P = 0·003), perhaps because areas with higher disturbance scores were mostly developed and there was less room for introduced grassland.

Disturbance PC scores did not correlate with standardized clutch initiation date (rs = 0·03, P = 0·82), indicating no relationship between disturbance and nest initiation. Additionally, standardized clutch initiation date did not correlate with habitat characteristics (agriculture: rs = −0·20, P = 0·08; sagebrush steppe: rs = 0·05, P = 0·65; introduced vegetation: rs = 0·06, P = 0·60), as might be predicted if different habitat characteristics had consistently higher or lower territory quality.

Disturbance PC scores and distance to road best explained reproductive outcome in American kestrels (Table 2). As disturbance PC scores increased, the probability of reproductive failure increased (β = 1·13, 85% CI: 0·52–1·74, Fig. 2). Nests near high disturbance areas (disturbance PC score > −0·78) were 9·9 times more likely to fail compared to nests in low disturbance areas (disturbance PC ≤ score −0·78). Probability of reproductive failure decreased as categorical nest distance from road increased (β = −2·63, 85% CI: −4·93 to −0·33, Fig. 2). The probability of failure increased more rapidly when nest box distance from road was ≤ 17·5 m from the centre of road when compared to those > 17·5 m but less than 34 m from a road. The effect of nest box distance from road on failure was less pronounced when nests were greater than or equal to 34 m from a road (Fig. 2). The other top 3 models contained disturbance PC and distance to road with another variable (Table 2); however, the effects of shrub-steppe (β = 0·00, 85% CI: −0·03 to 0·04) or standardized clutch initiation date (β = 1·09, 85% CI: −1·26 to 3·43) were lower and confidence intervals overlapped zero, indicating that these variables were less reliable predictors of outcome.

Table 2. Candidate models, ΔAICc, Akaike weight (wi), log-likelihood (LL) and number of parameters (K) for evaluating the role of human disturbance, habitat characteristics and clutch initiation date to explain American kestrel reproductive outcome in south-western Idaho in 2008 and 2009. Human disturbance was represented by a disturbance PC score and the distance between the nest box and the nearest road
ModelΔAICca w i LL K
  1. a

    Min AICc = 63·19.

  2. b

    Disturbance PC score representing the proportion of developed land within a 900-m radius circle centred on the nest box, and the number of traffic lanes, posted roadway speed and average number of vehicles per day on the road nearest the nest box.

  3. c

    The distance between the nest box and nearest road.

  4. d

    Standardized clutch initiation date (see 'Materials and methods').

  5. e

    The proportion of shrub-steppe vegetation within a kestrel territory.

Disturbance PCb + road_distc0·000·53−28·413
Disturbance PC + road_dist + clutch dated1·730·22−28·154
Disturbance PC + road_dist + shrub-steppee2·250·17−28·414
Disturbance PC + road_dist + shrub-steppe + clutch date4·060·07−28·145
Shrub-steppe22·690−40·852
Intercept-only model22·930−42·031
Shrub-steppe + clutch date23·070−39·953
Clutch date23·230−41·122
image

Figure 2. Probability of American kestrel reproductive failure in relation to disturbance principal component (PC) score and nest box distance from road: (a) distance ≤ 17·5 m, (b) distance > 17·5 m and < 34 m, (c) distance ≥ 34 m. Disturbance PC scores were based on traffic variables and the proportion of human development within a kestrel territory. Nests near larger, busier roads and developed landscapes received a higher disturbance PC score and nests near smaller, less busy roads and undeveloped landscapes received a lower disturbance PC score. As the disturbance PC score increased, the probability of failure increased; however, this relationship depended on the distance between the nest box and the nearest road. As distance from the road increased, the effect of the disturbance PC score on failure decreased. The solid line represents the predicted probability of failure based on disturbance PC score, dashed lines represent 85% confidence intervals, and circles show the disturbance PC scores for failed (1) or successful (0) nests.

Download figure to PowerPoint

Corticosterone Concentrations

Within a nesting pair, male and female CORT were not correlated (r2 = 0·001, P = 0·97). Disturbance PC scores were positively associated with higher female CORT (χ2 = 6·07, P = 0·01; β = 0·10, 95% CI 0·02–0·19), but there was no association between disturbance PC scores and male CORT (χ2 = 0·03, P = 0·86; β = −0·01, 95% CI: −0·08 to 0·07). In a comparison of nests that were successful with nests that were abandoned, the probability of reproductive failure increased as female CORT increased (χ2 = 4·37, P = 0·04; β = 1·09, 95% CI: 0·001–2·20, Fig. 3); however, there was no predictive relationship between male CORT and reproductive failure (χ2 = 0·41, P = 0·52; β = −0·43, 95% CI: = −1·75–0·89). Standardized clutch initiation date, the proportion of agriculture, sagebrush steppe or introduced grassland was not associated with female or male CORT (all P-values > 0·10). We found no effect of date (males: N = 54, r2 = 0·02, P = 0·26, females: N = 64, r2 = 0·02, P = 0·22) or time of day (males: r2 = 0·007, P = 0·55, females: r2 = 0·001, P = 0·79) on CORT.

image

Figure 3. Relationship between incubating American kestrel female baseline corticosterone concentrations (log pg mL−1) and the probability of reproductive failure. As female baseline corticosterone concentrations increased, the probability of failure increased. The solid line represents the predicted probability of failure based on female corticosterone, dashed lines represent 95% confidence intervals, and circles show the corticosterone concentrations of females that failed (1) or were successful (0).

Download figure to PowerPoint

Discussion

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

American kestrels are considered tolerant of human activity because of their widespread use of agricultural, rural-residential and ex-urban landscapes. Kestrels may use these landscapes because they provide favourable foraging conditions (e.g. short grass, open fields) and an abundance of nest sites (e.g. barns and old buildings). Despite kestrel presence in human-dominated landscapes, proximity to large, busy roads and developed areas negatively affected kestrel reproduction and acted as a stressor promoting nest abandonment. These results illustrate that while some species may be considered ‘tolerant’ of human landscapes, they may still be sensitive to (and negatively affected by) noxious stimuli, such as traffic noise (Francis, Ortega & Cruz 2011).

Human Disturbance and Reproductive Outcome

American kestrels nesting near busy roads were likely exposed to high noise levels (see Fig. 1 in Halfwerk et al. 2011). Traffic volume and vehicle speed, two of the variables we used to quantify human disturbance, are strong predictors of traffic noise (USDOT 1980). In their study of great tits Parus major (Linneaus 1758), Halfwerk et al. (2011) proposed four mechanisms why traffic noise may affect avian productivity: (i) noise interferes with acoustic assessment of mate quality, (ii) there is a non-random distribution of individuals along a noise gradient, with low-quality birds nesting in noisy areas, (iii) noise causes physiological stress, and (iv) noise impacts parent–offspring communication. Differences between kestrels and tits may provide insight into some of these mechanisms. For example, female kestrels select mates based on visual displays and provisioning (Johnsgard 1990) and do not rely on auditory communication to assess mate quality. Thus, traffic noise was unlikely to affect mate assessment. We found no relationship between disturbance PC score and clutch initiation date, an important predictor of local fitness in kestrels (K. Steenhof and J. A. Heath, unpublished data). This suggests that kestrels in disturbed areas were of similar quality to those in less disturbed areas and that noise disturbance may not affect kestrel nest site selection. Finally, kestrel nest abandonment occurred during egg incubation, indicating that noise interference with parent–offspring communication was not likely to explain kestrel reproductive failure.

Our results support the hypothesis that increased noise causes physiological stress (also see 'Discussion' section on CORT). Noise may impact prey detection and foraging ability (Schaub, Ostwald & Siemers 2008) or, for cavity-nesting species that rely on acoustic signals as part of their sensory ecology, anthropogenic noise may mask auditory cues associated with predator detection (Quinn et al. 2006). Nesting kestrels are vulnerable to mammalian predators that block the cavity entrance before the adult can escape (J. A. Heath, unpublished data). Kestrels in our study flushed from their nest in response to the sound of approaching footsteps or alarm calls given by ground squirrels (Spermophilus spp.). Cavity nesters in noisy environments may compensate for decreased auditory cues by increasing vigilance behaviour, such as visual scans from the nest entrance or flushing from the nest, leading to changes in energy allocation or extended periods away from the nest during incubation. Future experimental studies that manipulate sound within listening areas (Barber, Crooks & Fristrup 2010) around nests and examine trade-offs between vigilance behaviour and parental care will help to determine the proximate mechanism leading to reproductive failure.

An alternative explanation for why kestrels nesting near large, busy roads and developed areas had higher rates of nest failure could be that kestrels at these locations were more likely to be hit by vehicles. We found evidence (three carcasses at failed nests) that vehicular collisions accounted for a small proportion of nest failure. It seems unlikely that detection bias lead to an underestimate of vehicle collision rate, given the high probability of carcass detection and low rate of scavenger activity in our quail trials (Strasser 2010). Although vehicle collisions may not have been the mechanism leading to nest failure in this study, collisions with anthropogenic structures such as vehicles, buildings and power lines are a significant cause of avian mortality (Erickson, Johnson & Young 2005). The effects of vehicle collisions on adult (and fledgling) survival are an important research direction for understanding the impacts of roads on wildlife populations.

Corticosterone Concentrations

Female kestrels in areas with higher disturbance PC scores had higher baseline CORT concentrations and were more likely to abandon nests than females in areas with lower disturbance PC scores. These results support the hypotheses that disturbance from traffic and human land use acts as a stressor and that increases in CORT can lead to reproductive failure and potential population-level impacts. Over time, the cumulative effect of elevated baseline CORT may lead to changes in physiology and behaviour that redirects energy away from reproduction towards survival, resulting in nest abandonment. This physiological mechanism is consistent with the result that kestrels select nest sites in high disturbance areas yet have high rates of abandonment after egg laying. Given the relatively rapid increase in anthropogenic noise, organisms may not perceive noise as an indicator of poor habitat quality and instead rely on cues, such as foraging areas, to evaluate habitat quality. Studies that are designed to understand whether traffic conditions may create ecological traps would benefit from a design that compared CORT and abandonment across species that differed in their use of listening areas for predator detection (e.g. cavity nesters versus open-cup nesters).

Male kestrels in high disturbance areas did not have elevated CORT, and male CORT was not predictive of nest abandonment. These results are consistent with other studies that have found sex-specific responses to stressors and reproductive outcome (Bonier et al. 2007). Sex-specific relationships between CORT, disturbance and outcome could reflect sex-specific differences in the amount of time males and females spend incubating. For example, female kestrels spent more time incubating and may have had higher exposure to stressors resulting in elevated CORT and nest abandonment. Alternatively, the breeding ecology of birds may lead to sex-specific decisions about nest abandonment. Future studies on disturbance, baseline CORT and reproductive outcomes should consider sex-specific exposure, energetic investment and future reproductive potential.

Conservation Concerns in Human-Dominated Landscapes

Species presence in a human-dominated landscape does not equate to a tolerance for anthropogenic stressors across all stages of the annual cycle, and more specifically, managers should not discount human disturbance as a potential cause of kestrel population declines. Although some (6%) kestrel pairs renested after abandonment, later clutch initiation is associated with reduced local fitness (K. Steenhof and J. A. Heath, unpublished data) and many pairs may delay breeding until a subsequent year. For species with relatively short reproductive lifetimes (Steenhof & Heath 2009), reproductive failure in any 1 year can have strongly negative repercussions for lifetime reproductive success (Newton 2003) and, eventually, population size. Further, human disturbance was more predictive of reproductive outcome than other factors commonly associated with population size (habitat) and reproductive success (clutch initiation date).

The negative effects of human disturbance on animal physiology, reproduction and survival will become more pervasive as roads, development and recreational activities spread across the landscape. There is increasing evidence that humans significantly alter the way organisms perceive their environment, specifically via changes to the soundscape (Barber, Crooks & Fristrup 2010; Halfwerk et al. 2011; Bennett & Zurcher 2013). Future research that measures the acoustic environment while documenting physiological and behavioural responses may be able to identify species-specific noise thresholds (Bennett & Zurcher 2013) and lead to a better understanding of the noise levels that cause population-level impacts. In protected areas, managers may implement techniques to reduce noise, such as sound barriers around roads or regulations that decrease traffic volume and speed (Halfwerk et al. 2011 and references within; Kight, Saha & Swaddle 2012). However, site-specific management would be an impracticable option for a widespread species like the American kestrel. We suggest that the most effective management technique to mitigate for noise-related effects on wildlife will be regulations or economic incentives that encourage engineering innovations that result in quieter roads, vehicles and communities. Until such technology is realized, managers should carefully consider or discourage projects that juxtapose favourable habitat conditions with areas of high human activity (i.e. busy roads) to decrease risk of ecological traps.

Acknowledgements

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

This project was supported by Boise State University, the Raptor Research Center, North American Bluebird Society, the Society for Integrative and Comparative Biology, Boise State University's Provost Office undergraduate research work study funds and the NSF Idaho EPSCoR Program (EPS-0814387). All procedures were conducted with approval from Boise State University's Institutional Animal Care and Use Committee (protocol #006-08-007). We appreciate the Idaho land owners and the Idaho Department of Transportation that allowed us to access kestrel boxes on their property. We thank M. Foster, C. Hayes, A. Webber, D. Owen and T. Patel for help in the field and S. Alsup for help with GIS. A.M. Dufty, Jr., C.A. Lott, S.J. Novak, E. Fernández-Juricic, D. Thompson and two anonymous reviewers made comments that improved earlier drafts of this manuscript. This paper is dedicated to A.M. Dufty, Jr.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Antworth, R.L., Pike, D.A. & Stevens, E.E. (2005) Hit and run: effects of scavenging on estimates of road killed vertebrates. Southeastern Naturalist, 4, 647656.
  • Arnold, T.W. (2010) Uninformative parameters and model selection using Akaike's Information Criterion. Journal of Wildlife Management, 74, 11751178.
  • Barber, J.R., Crooks, K.R. & Fristrup, K.M. (2010) The costs of chronic noise exposure for terrestrial organisms. Trends in Ecology & Evolution, 25, 180189.
  • Baudains, T.P. & Lloyd, P. (2007) Habituation and habitat changes can moderate the impacts of human disturbance on shorebird breeding performance. Animal Conservation, 10, 400407.
  • Beale, C.M. & Monaghan, P. (2004) Human disturbance: people as predation-free predators? Journal of Applied Ecology, 41, 335343.
  • Bennett, V.J. & Zurcher, A.A. (2013) When corridors collide: road-related disturbance in commuting bats. The Journal of Wildlife Management, 77, 93101.
  • Bonier, F., Martin, P.R., Sheldon, K.S., Jensen, J.P., Foltz, S.L. & Wingfield, J.C. (2007) Sex-specific consequences of life in the city. Behavioral Ecology, 18, 121129.
  • Burnham, K.P. & Anderson, D.R. (2002) Model Selection and Multimodel Inference: A Practical Information-Theoretic Approach, 2nd edn. Springer-Verlag, New York, NY.
  • Busch, D.S. & Hayward, L.S. (2009) Stress in a conservation context: a discussion of glucocorticoid actions and how levels change with conservation-relevant variables. Biological Conservation, 142, 28442853.
  • Creel, S., Fox, J.E., Hardy, A., Sands, J., Garrott, B. & Peterson, R.O. (2002) Snowmobile activity and glucocorticoid stress responses in wolves and elk. Conservation Biology, 16, 809814.
  • Cyr, N.E. & Romero, L.M. (2007) Chronic stress in free-living European starlings reduces corticosterone concentrations and reproductive success. General and Comparative Endocrinology, 151, 8289.
  • Ellenberg, U., Setiawan, A.N., Cree, A., Houston, D.M. & Seddon, P.J. (2007) Elevated hormonal stress response and reduced reproductive output in yellow-eyed penguins exposed to unregulated tourism. General and Comparative Endocrinology, 152, 5463.
  • Environmental Systems Resource Institute (2009) ArcMap 9.2. ESRI, Redlands, CA.
  • Erickson, W.P., Johnson, G.D. & Young, D.P. (2005) A summary and comparison of bird mortality from anthropogenic causes with an emphasis on collision. USDA Forest Service General Technical Report PSWGTR, 191, 10291042.
  • Farmer, C.J. & Smith, J.P. (2009) Migration monitoring indicates widespread declines of American kestrels (Falco Sparverius) in North America. Journal of Raptor Research, 43, 263273.
  • Fernández, C. & Azkona, P. (1993) Human disturbance affects parental care of marsh harriers and nutritional-status of nestlings. Journal of Wildlife Management, 57, 602608.
  • Francis, C.D., Ortega, C.P. & Cruz, A. (2011) Noise pollution filters bird communities based on vocal frequency. Public Library of Science ONE, 6, e27052.
  • French, S.S., González-Suárez, M., Young, J.K., Durham, S. & Gerber, L.R. (2011) Human disturbance influences reproductive success and growth rate in California sea lions (Zalophus californianus). Public Library of Science ONE, 6, e17686.
  • Frid, A. & Dill, L.M. (2002) Human-caused disturbance stimuli as a form of predation risk. Conservation Ecology, 6, 11. http://www.ecologyandsociety.org/vol6/iss1/art11/.
  • Gill, J.A., Sutherland, W.J. & Watkinson, A.R. (1996) A method to quantify the effects of human disturbance on animal populations. Journal of Applied Ecology, 33, 786792.
  • Halfwerk, W., Holleman, L.J.M., Lessells, C.M. & Slabbekoorn, H. (2011) Negative impact of traffic noise on avian reproductive success. Journal of Applied Ecology, 48, 210219.
  • Harvey, S., Phillips, J.G., Rees, A. & Hall, T.R. (1984) Stress and adrenal-function. Journal of Experimental Zoology, 232, 633645.
  • Johnsgard, P. (1990) Hawks, Eagles, and Falcons of North America. Smithsonian Press, Washington, DC.
  • Kight, C.R., Saha, M.S. & Swaddle, J.P. (2012) Anthropogenic noise is associated with reductions in the productivity of breeding eastern bluebirds (Sialia sialis). Ecological Applications, 22, 19891996.
  • Manly, B.F.J. (1994) Multivariate Statistical Methods: A Primer, 2nd edn. Chapman and Hall/CRC, Washington, DC.
  • Miller, D.A., Vleck, C.M. & Otis, D.L. (2009) Individual variation in baseline and stress-induced corticosterone and prolactin levels predicts parental effort by nesting mourning doves. Hormones and Behavior, 56, 457464.
  • Newton, I. (2003) Population Limitation in Birds. Academic Press, San Diego, CA.
  • Papouchis, C.M., Singer, F.J. & Sloan, W.B. (2001) Responses of desert bighorn sheep to increased human recreation. The Journal of Wildlife Management, 65, 573582.
  • Parris, K.M. & Schneider, A. (2008) Impacts of traffic noise and traffic volume on birds of roadside habitats. Ecology and Society, 14, 29. http://www.ecologyandsociety.org/vol14/iss1/art29/.
  • Quinn, J.L., Whittingham, M.J., Butler, S.J. & Cresswell, W. (2006) Noise, predation risk compensation and vigilance in the chaffinch Fringilla coelebs. Journal of Avian Biology, 37, 601608.
  • Rodríguez-Prieto, I. & Fernández-Juricic, E. (2005) Effects of direct human disturbance on the endemic Iberian frog Rana iberica at individual and population levels. Biological Conservation, 123, 19.
  • Rohrbaugh, R.W. Jr & Yahner, R.H. (1997) Effects of macrohabitat and microhabitat on nest-box use and nesting success of American kestrels. The Wilson Bulletin, 109, 410423.
  • Romero, L.M. & Reed, J.M. (2005) Collecting baseline corticosterone samples in the field: is under 3 min good enough? Comparative Biochemistry and Physiology a-Molecular and Integrative Physiology, 140, 7379.
  • SAS Institute (2002) SAS User's Guide, Version 9.2. SAS Institute, Inc., Cary, NC.
  • Schaub, A., Ostwald, J. & Siemers, B.M. (2008) Foraging bats avoid noise. Journal of Experimental Biology, 211, 31743180.
  • Schlaepfer, M.A., Runge, M.C. & Sherman, P.W. (2002) Ecological and evolutionary traps. Trends in Ecology & Evolution, 17, 474480.
  • Schlesinger, M.D., Manley, P.N. & Holyoak, M. (2008) Distinguishing stressors acting on land bird communities in an urbanizing environment. Ecology, 89, 23022314.
  • Silverin, B. (1998) Behavioural and hormonal responses of the pied flycatcher to environmental stressors. Animal Behaviour, 55, 14111420.
  • Smallwood, J.A. & Bird, D.M. (2002) American kestrel (Falco sparverius). The Birds of North America Online (ed. A. Poole), pp. 230. Cornell Lab of Ornithology, Ithaca, NY.
  • Steenhof, K. & Heath, J.A. (2009) American kestrel reproduction: evidence for the selection hypothesis and the role of dispersal. Ibis, 151, 493501.
  • Steenhof, K. & Peterson, B.E. (2009) American kestrel reproduction in southwestern Idaho: annual variation and long-term trends. Journal of Raptor Research, 43, 283290.
  • Strasser, E.H. (2010) Reproductive failure and the stress response in American kestrels nesting along a human disturbance gradient. Master's Thesis, Boise State University.
  • Strasser, E.H. & Heath, J.A. (2011) Effects of developmental conditions on nestling American kestrel (Falco sparverius) corticosterone concentrations. General and Comparative Endocrinology, 173, 164170.
  • Thiel, D., Jenni-Eiermann, S., Braunisch, V., Palme, R. & Jenni, L. (2008) Ski tourism affects habitat use and evokes a physiological stress response in capercaillie Tetrao urogallus: a new methodological approach. Journal of Applied Ecology, 45, 845853.
  • U.S. Department of Transportation (1980) Highway Traffic Noise Report No. HEV-21/8-80 (20M), Federal Highway Administration, Washington, DC.
  • Webber, A.F., Heath, J.A. & Fischer, R.A. (2013) Human disturbance and stage-specific habitat requirements influence snowy plover site occupancy during the breeding season. Ecology and Evolution, 3, 853863. DOI:10.1002/ece3.511.
  • Weilgart, L.B. (2007) The impacts of anthropogenic ocean noise on cetaceans and implications for management. Canadian Journal of Zoology, 85, 10911116.
  • Williams, R., Lusseau, D. & Hammond, P.S. (2006) Estimating relative energetic costs of human disturbance to killer whales (Orcinus orca). Biological Conservation, 133, 301311.
  • Wingfield, J.C., Vleck, C.M. & Moore, M.C. (1992) Seasonal changes of the adrenocortical response to stress in birds of the Sonoran Desert. Journal of Experimental Zoology, 264, 419428.
  • Wingfield, J.C., Maney, D.L., Breuner, C.W., Jacobs, J.D., Lynn, S., Ramenofsky, M. & Richardson, R.D. (1998) Ecological bases of hormone-behavior interactions: the “emergency life history stage”. American Zoologist, 38, 191206.