BioOne.org will be down briefly for maintenance on 17 December 2024 between 18:00-22:00 Pacific Time US. We apologize for any inconvenience.
Open Access
How to translate text using browser tools
8 June 2021 Temporal Genetic Diversity and Effective Population Size of the Reintroduced Aplomado Falcon (Falco femoralis) Population in Coastal South Texas
Jeff A. Johnson, Alexandra Stock, Paul Juergens, Brian Mutch, Christopher J.W. McClure
Author Affiliations +
Abstract

Reintroductions are an important tool in conservation for preserving and enhancing biodiversity and preventing extinction, and post-release monitoring is essential to evaluate and inform conservation management and maximize recovery success. By quantifying genetic diversity levels and effective population size (Ne) over time, managers can gauge to what degree additional efforts are needed to increase the likelihood of population persistence. The endangered Northern Aplomado Falcon (Falco femoralis septentrionalis) population in South Texas was reestablished and supplemented with captive-bred individuals originating from 27 founders collected in eastern Mexico (San Luis Potosí, Veracruz, Tabasco, and Chiapas). A total of 927 Aplomado Falcons were released at 23 locations along the southern coast of Texas between 1985 and 2004, and in 2012 and 2013. To assess the species' reintroduction and recovery, we applied a genetic monitoring approach using sampled nestlings (n = 267) from a total of 108 nests in 2004–2005 and 2012–2016. Based on ten microsatellite loci, levels of genetic diversity (i.e., allelic richness and heterozygosity) remained stable over the sampled time period, with no indication of inbreeding. Diversity levels were comparable to a subset of samples collected from the captive population founders (n = 11). Similarly, individuals from the South Texas population showed strong admixture with the founding population, and levels of both Ne and of effective breeding (Nb) showed no signs of decline over the sampled time period. To what degree overlapping generations and the release of additional Aplomado Falcons during the sampled time period limited our assessment of the South Texas population is not fully known. Continued monitoring across multiple generations is advisable to assess the population's ability to persist.

Introduction

Genetic diversity allows populations to adapt to changing environments and reduces potential for inbreeding depression (Allendorf et al. 2012, Lai et al. 2019). Management of genetic diversity is therefore an important component of increasing the likelihood of recovery of threatened and endangered species (Harrisson et al. 2014, Willoughby et al. 2015, Funk et al. 2019). Ensuring adequate levels of genetic diversity is especially important when using translocated or captive-reared individuals to establish a new population. Estimates of effective population size (Ne) as originally described by Wright (1931, 1938) provide a measure for gauging the likelihood of a population's ability to maintain genetic diversity over time (Hare et al. 2011, Jamieson and Allendorf 2012), and can be used as a practical measure for the conservation and management of wild populations (Schwartz et al. 2007). Ne represents the size of an idealized panmictic population where each adult has an equal expectation of generating viable offspring with the same gene frequency drift or inbreeding as the observed population (Wright 1931), or simply the number of breeding adults within a population. Because it is not always fully known if all released individuals contribute to the breeding population, managers can assess to what degree additional efforts are required to mitigate further genetic diversity loss by monitoring Ne over time.

Ongoing conservation efforts focused on reestablishing a self-sustaining Northern Aplomado Falcon (Falco femoralis septentrionalis) population in the United States of America (USA) would benefit from knowledge of levels of genetic diversity and estimates of Ne over time compared to the Northern Aplomado Falcon founding population. The Northern Aplomado Falcon (hereafter Aplomado Falcon) historically inhabited areas in the Texas Coastal Plains and the Chihuahuan Desert in West Texas, southern New Mexico, and southeastern Arizona. The subspecies has declined significantly in abundance over the past century, with the last breeding pair in the USA (prior to captive release efforts initiated in the mid-1980s) observed in the wild in 1952 (Hunt et al. 2013). Although the species' overall geographic distribution extends south through Mexico to Argentina, the distribution of the most northern subspecies (F. f. septentrionalis) is from Guatemala to the southern USA (Keddy-Hector 2019, Keddy-Hector et al. 2020). This subspecies has been identified by the US Fish and Wildlife Service (USFWS) as federally endangered (51 FR 6686) since 1986.

To reestablish the Aplomado Falcon in the USA, 25 wild nestlings (from at least 15 different nests) and two adults donated by falconers (one originating from a nest in Tabasco and the other of unknown origin) were obtained between 1977 and 1988 from the four Mexican States of San Luis Potosí, Veracruz, Tabasco, and Chiapas (see  Supplemental Material 1 (169_rapt-55-02-09_s01.pdf)). The captive population was created with the intention to propagate and release captive-hatched individuals in an effort to restore the species to its original range within the southern USA. The first captive-hatched Aplomado Falcons were produced in 1982 (Cade et al. 1991) and the first were released to the wild in 1985 (Hunt et al. 2013). All released Aplomado Falcons were offspring of the 27 founders or their subsequent generations that were reared and bred in captivity.

Although substantial effort was made in an attempt to establish a breeding population in West Texas and New Mexico, only one nesting pair was observed in New Mexico after 2012 despite the release of 637 captive-hatched Aplomado Falcons in West Texas between 2002 and 2011, and more than 337 in New Mexico between 2006 and 2012 (Hunt et al. 2013, USFWS 2014). The exact reason the released Aplomado Falcons failed to establish is not fully known, but multiple years of severe drought and deterioration of ground cover likely decreased prey availability, which in turn likely decreased the population's productivity during the reintroduction efforts (Hunt et al. 2013, Sweikert and Phillips 2015). In contrast, the release efforts in coastal South Texas have been more successful. A total of 927 Aplomado Falcons were released at 23 locations along the southern coast of Texas between 1985 and 2004, and in 2012 and 2013 (Jenny et al. 2004, Hunt et al. 2013). These resulted in the establishment of two general breeding areas. Between 2008 and 2013, 15 to 20 mated pairs were observed each year north of Brownsville (herein identified as Laguna Atascosa National Wildlife Refuge, LANWR) and 13 to 17 pairs on Matagorda Island, San Jose Island, and Mustang Island near Rockport, Texas (herein identified as Matagorda Island National Wildlife Refuge, MINWR; Hunt et al. 2013). Since 2013, the South Texas population grew annually through 2017, when 39 territorial pairs were identified within the LANWR and MINWR populations.

Nestling survival within the South Texas population appeared to have benefited greatly from the use of raised artificial nest platforms with barred boxes over the nest that prevent predation of young by Great Horned Owls (Bubo virginianus), the primary predator of Aplomado Falcons (Jenny et al. 2004, Hunt et al. 2013). Reproductive rates in 2012 and 2013 for the LANWR population averaged 1.60 and 1.93 fledglings per nesting attempt, respectively, and 1.85 and 2.00 fledglings per nesting attempt, respectively, for the MINWR population (Hunt et al. 2013). Current efforts are focused on establishing additional breeding territories in South Texas, with increased attention on habitat restoration and installation of additional artificial nest platforms (e.g., McClure et al. 2017a); there are no immediate plans to release additional captive-bred falcons. Other than the number of known breeding pairs at artificial nest platforms, no information is available concerning the South Texas population's effective population size (Ne) and its genetic variability compared to its founding population.

Post-release demographic monitoring of translocated and reestablished populations is essential to ensure population persistence (e.g., Ewen and Armstrong 2007, Seddon et al. 2007, McClure et al. 2017b). Here we applied a population genetic approach using multi-locus genotypic data generated from microsatellite DNA to monitor temporal trends in population genetic diversity and effective population size (Ne) of the Aplomado Falcon population in South Texas between 2004 and 2016. We predicted that levels of genetic diversity and Ne of the South Texas Aplomado Falcon population were similar to those of the founding population, given the recency with which individuals from the captive propagation program were released. The results from this study provide a baseline measure for future monitoring efforts to help gauge progress toward achieving a viable self-sustaining Aplomado Falcon population in the USA.

Methods

Tissue Collection and DNA Extraction. We obtained muscle tissue from 11 of the 27 founders of the captive Aplomado Falcon population. The carcasses were kept frozen (–20°C) prior to sampling for this study. For the South Texas population, we obtained blood samples from a total of 267 Aplomado Falcon nestlings (11 to 18 nests per year between 2004 and 2016) in two coastal breeding areas, LANWR and MINWR (Table 1). We extracted DNA from all samples using the DNeasy Blood and Tissue Kit following manufacturer's protocols (Qiagen, Inc., Germantown, MD, USA).

Table 1.

Number of Aplomado Falcon samples collected by year and location within South Texas (numbers in parentheses indicate the number of nests sampled). Laguna Atascosa National Wildlife Refuge, LANWR; Matagorda Island National Wildlife Refuge, MINWR.

img-z3-5_169.gif

Microsatellite DNA Genotyping. We screened a total of 19 microsatellite loci previously described for Orange-breasted Falcon (F. deiroleucus; Beasley et al. 2014) for polymorphism in Aplomado Falcon using PCR methods as described elsewhere (Beasley et al. 2014). We genotyped samples using an ABI 3130xl Genetic Analyzer (Applied BioSystems, Foster City, CA, USA) and the program GeneMarker v.1.6 (SoftGenetics, LLC, State College, PA, USA). Based on a subset of ten unrelated individuals, ten microsatellite loci (Fade-10, Fade-13, Fade-14, Fade-15, Fade-16, Fade-20, Fade-22, Fade-31, Fade-33, Fade-48) possessed multiple alleles in Aplomado Falcon, resulting in unambiguous allele calls. We genotyped all remaining samples with the ten polymorphic loci.

Pairwise Relatedness. Data sets containing a high proportion of close relatives can influence the ability to obtain accurate estimates of genetic diversity and population structure (Rodriguez-Ramilo and Wang 2012, de Jager et al. 2017, but see Waples and Anderson 2017). To identify unrelated individuals, we calculated pairwise genetic estimates of relatedness among all sampled individuals based on Wang's relatedness coefficient (rw; Wang 2002) with the R package related (Pew et al. 2015; see also Wang 2011) in R v.3.4.0 (R Core Team 2017). We conducted simulation analyses to determine which of seven different methods for calculating relatedness with coancestry was most appropriate given the characteristics of the generated microsatellite data set (Csillery et al. 2006, Taylor 2015). We simulated data sets of 100 pairs for four relatedness categories (parent–offspring, full-sibling, half-sibling, and unrelated), with allele frequencies calculated from the study population's allele frequencies. We then used seven estimators to calculate relatedness, and compared the estimates from the simulations to those expected based on diploid inheritance patterns (parent-offspring = 0.50, full-siblings = 0.5, half-siblings = 0.25, and unrelated = 0.0) using their correlation coefficients. The Wang (2002) estimator was identified with the highest correlation coefficient (Pearson's r = 0.83), and used in subsequent analyses (see also  Supplemental Material 2, Fig. S1 (169_rapt-55-02-09_s02.pdf)).

All nestling samples obtained from the same nest should indicate full-sibling relationships based on their pairwise relatedness value (r = 0.5) when assuming monogamy, whereas half-sibling relationships or extra-pair young (EPY; r = 0.25) may occur due to extra-pair paternity (EPP) when young have different sires or when intraspecific brood parasitism occurs. Our ability to accurately estimate relatedness between individuals using genetic data, however, is dependent on a number of factors including the number and level of polymorphism of the microsatellite loci (Van de Casteele et al. 2001, Blouin 2003, Csillery et al. 2006). Considerable sampling variance often exists around point-estimates of relatedness using microsatellite genotypic data (e.g.,  Supplemental Material 2, Fig. S1 (169_rapt-55-02-09_s02.pdf)), and therefore, measures must be taken to account for the potential uncertainty (Taylor 2015).

We compared pairwise relatedness values for all young within nests surveyed in South Texas to verify full-sibling relationships (rw ≈ 0.5) based on their 95% CIs calculated using bootstrapping over loci (1000 replicates). For cases where young from the same nest possessed upper 95% CI estimates <0.45, we attempted to identify nests within the same subpopulation (LANWR or MINWR) with young that could be half-siblings (rw ≈ 0.25). All young within the nest would also need to show a similar relationship with the other young, and shared paternity for nest sites within close geographic proximity was assumed more likely than for nests farther away. When more than three nestlings were present in a nest, we also determined whether allele combinations would allow us to reject full-sibling relationship based on diploid inheritance pattern. Using the above criteria, we feel confident that the young identified as half-siblings within the same nest represent the minimum number of EPY cases within the population.

Genetic Diversity. We created two data sets to assess choice of method for reducing relatedness among samples prior to calculating genetic diversity metrics. One data set consisted of a single individual per nest per year, and the second data set was created using the program Friends-and-Family (de Jager et al. 2017) with a relatedness cutoff value of 0.45 based on the Wang (2002) estimator. Although de Jager et al. (2017) recommended a cutoff value of 0.25, that parameter resulted in insufficient sample sizes for population genetic analyses. Instead, we follow recommendations of Waples and Anderson (2017) who showed that using a less stringent cut-off value of 0.5 could result in more reliable output in population genetic analyses.

For each data set, we tested microsatellite genotypes for linkage disequilibrium and departure from Hardy-Weinberg Equilibrium (HWE) for each study site (founders, LANWR, and MINWR) and locus with the program GDA v. 1.1 (Lewis and Zaykin 2001). We corrected for multiple simultaneous comparisons using sequential Bonferroni corrections (Rice 1989). We also calculated total number of alleles (A), allelic richness (AR), mean observed (Ho) and expected heterozygosity (He), and inbreeding coefficient (f ) values using the R package DIVERSITY v.1.9.89 (Keenan et al. 2013) in R v.3.3.2 (R Core Team 2016). We calculated AR to control for differences in the number of alleles among populations that differed in sample size (Leberg 2002), and constructed 95% CIs by means of 1000 bootstrap replications to estimate statistical significance of AR estimates between populations and temporal periods. We considered values for f as significant if their 95% CI did not overlap with zero.

Population Genetic Differentiation. We used principal coordinate analysis (PCoA) implemented in the Excel-based genetic analysis program GenAlEx v. 6.503 (Peakall and Smouse 2012) to visualize whether any pattern existed among individual samples that corresponds with temporal sampling period and study site location. We produced a standardized genetic distance PCoA using pairwise genetic distances among all individuals with the codom-genotypic option as implemented in GenAlEx (Peakall and Smouse 2012). The genetic distances were based on the number of alleles shared by individuals and their respective heterozygosity level (Smouse and Peakall 1999).

To assess genetic differentiation among temporal sampling periods and study sites, we used the Bayesian clustering program Structure v.2.3.4 (Pritchard et al. 2000) to identify the most likely number of clusters (K) and assign individuals to the inferred population cluster(s). We used an admixture model with correlated allele frequencies and a burn-in of 2.5×105 followed by 5×105 permutations for K = 1 to 10 with 10 iterations for each K. We conducted analyses without the LOCPRIOR model. We calculated the membership coefficient (q), which represents a measure of proportional individual membership to different inferred population clusters allowing an assessment of admixture from two or more population clusters. We used the web-based program Structure Harvester v. A.1 (Earl and vonHoldt 2012) and the Evanno ΔK method (Evanno et al. 2005) to determine the mostly likely number of genetically distinct clusters (K). We used the program CLUMPP v. 1.1.2 (Jakobsson and Rosenberg 2007) to compile replicate run results from Structure, and used the program STRUCTURE PLOT v.2.0 to visualize the results (Ramasamy et al. 2014).

We also used GenAlEx to calculate pairwise estimates of population differentiation based on FST and Jost's D (Dest; Jost 2008) among temporal sampling periods and study sites. Dest has been shown to be an effective estimator for describing allelic differentiation among populations and less affected by a DNA marker's level of polymorphism, which can limit maximum differentiation estimates as shown with FST (Jost 2008, Meirmans and Hedrick 2011). We used 1000 random permutations to test for departure from the null hypothesis of no genetic differentiation (Peakall and Smouse 2012).

Effective Population Size. We calculated estimates of effective number of breeders (Nb) and contemporary effective population size (Ne) for each sampled time period and study site, respectively. Nb is fairly quick to assess because only one season (or cohort) of data collection is required, allowing managers to compare the number of breeders that produced each cohort across multiple years to gauge breeding stability over time. In contrast, estimates of Ne represent the harmonic mean size of a breeding population over a single generation where fluctuations in size are less obvious. Estimates of Ne provide a valuable measure for conservation because the magnitude of Ne influences the rate of random genetic drift, loss of genetic diversity, and effectiveness of selection and migration (Charlesworth 2009). Our ability to estimate Ne accurately with wild populations, however, can be complicated due to overlapping generations with individuals that reproduce multiple times during their lifespan (Luikart et al. 2010, Waples et al. 2014).

Our sampling design allowed us to assess Ne for each of the study populations and Nb for each consecutive breeding cohort or year (from 2004–2005 and 2012–2016) within the South Texas population. To estimate Nb, we used two single-sample estimator methods that differed based on their use of sib-relationships among samples. The first method was the sibship assignment maximum likelihood method described in Wang (2009) that estimates Nb from frequencies of full- and half-sibling dyads from randomly sampled sets of offspring. This method has been shown to provide accurate estimates of Nb of the parent generation that produced the cohort, if samples are collected from a single cohort at the same life stage (e.g., offspring prior to fledging; Wang 2009, 2016). We estimated Nb(sib) and 95% CIs for each cohort using the program Colony2 (Jones and Wang 2010) with the “very long” duration option under the full likelihood model with medium precision and a uniform prior for sibship size while allowing for polygamous females (see Results).

The second method we used to estimate Nb was the linkage disequilibrium (LD) method as implemented in LDNe v. 1.31 (Waples and Do 2008) using the program NeEstimator v2 (Do et al. 2014). The LD method estimates Nb (or Ne depending on timing of sampling) based on measuring the nonrandom association of alleles at different loci (Hill 1981). Similar to the sibship assignment method, we obtained Nb(LD) estimates for each individual reproductive cohort, but we used a data set based on a single randomly selected individual per nest since the method assumes random mating. We excluded alleles with frequencies <0.02 (Waples 2006), and obtained 95% CIs using the jackknife option (Waples and Do 2008). We applied a correction to each estimate of Nb(LD) to address potential bias due to age-structured populations with overlapping generations (Waples et al. 2013, 2014) based on two life-history parameters corresponding with adult lifespan (AL = 15) and age at maturity (α = 1; see Table 3 in Waples et al. 2014) using banding records obtained directly from the South Texas population (B. Rolek pers. comm.). We also used the same life-history parameters to estimate Ne(LD-Adj2) for each Nb(LD-Adj2) estimate using methods described in Waples et al. (2014), and calculated a separate Ne(LD-pool) estimate after pooling all samples into a single data set.

Table 3.

Estimates of effective number of breeders (Nb) and population size (Ne) for the South Texas Aplomado Falcon population over time based on the sibship assignment (sib) and linkage disequilibrium (LD) methods. Estimates of Nb and Ne using LD have been adjusted to correct for biases due to age structure (see Table 3 in Waples et al. 2014) and 95% CIs are provided in parentheses.

img-z8-2_169.gif

Results

Pairwise Relatedness. Five percent (3 of 55) of the pairwise relatedness estimates (rw) among founders were ≥0.25 (i.e., relationships equivalent to half-siblings and above). One pairwise comparison indicated a full-sibling relationship (rw ≈ 0.5), and the other two were half-siblings (rw ≈ 0.25). Approximately 13% (3856 of 30,381) of the pairwise comparisons among all sampled individuals from the two management areas in South Texas (LANWR and MINWR) possessed rw ≥ 0.25. Mean rw among management areas was 0.085 (SD = 0.136). The percent of pairwise comparisons with rw ≥ 0.25 (or the equivalent of half-siblings, grandparent-grandnestling, aunt/uncle-nephew/niece and higher) between the sampled founders and among the temporal South Texas population ranged from 5.2% in 2013 to 8.8% in 2012 with no consistent pattern observed over time (Fig. 1).

Figure 1.

Percent of pairwise relatedness estimates (rw) between Aplomado Falcon founders and each of the temporal South Texas Aplomado Falcon breeding areas (LANWR and MINWR combined) that were ≥ 0.25 (or half-sibling, grandparent-grandnestling, aunt/uncle-nephew/niece and higher order relationships).

img-z6-5_169.jpg

Among the 108 sampled nests in South Texas, we failed to reject full-sibling relationships (rw ≈ 0.5) for the majority of young within a nest. However, at least seven nests (n = 2, 2004, LANWR and MINWR; n = 2, 2005, LANWR and MINWR; n = 2, 2013, LANWR; n = 1, 2016, LANWR) possessed relatedness estimates suggesting half-sibling relationships, three of which were confirmed based on diploid allele inheritance patterns (LANWR-2004, LANWR-2013, LANWR-2016), suggesting extra-pair paternity or intraspecific brood parasitism. For six of the seven nests with young that had half-sibling relationships, nestlings in a neighboring nest within the same management area were also identified as their half-siblings (rw ≈ 0.25). An additional eight nests held nestlings that may also have been half-siblings, but their upper 95% CI estimates of relatedness among within-nest comparisons were 0.45 ≤ rw ≤ 0.50.

We were able to investigate turnover of breeders at nest sites for 38 cases that included young sampled from the same nest site in at least two consecutive years over the duration of this study (2004–2005 and 2012–2016). Pairwise rw values among young from 31 of those 38 cases (82%) failed to reject that the same breeding pair occupied the same nest site over consecutive years (i.e., all young rw ≥ 0.30). In the six cases where nestlings were sampled from the same nest site for three consecutive years, pairwise rw values between years suggested they were all occupied by the same breeding pair for each nest site. Overlapping generations within the study population and insufficient precision in our estimates of rw to distinguish between half- and full-sibling relationships combined to preclude confident identification of breeding pairs that may have moved to different nest site locations between sampled years.

Genetic Diversity. All loci per population and temporal period were in Hardy-Weinberg Equilibrium after adjusting the significance level due to multiple comparisons (Rice 1989). Genetic diversity measures and population mean relatedness values were similar between the founders and the South Texas temporal sampled cohort population, and no significant change in diversity was observed over time (Table 2). Similar results were obtained with the data set that included only unrelated individuals as determined using Friends-and-Family method ( Supplemental Material 2, Table S1 (169_rapt-55-02-09_s02.pdf)).

Table 2.

Microsatellite DNA (10 loci) diversity estimates for each population temporal period. Estimates based on data set 1 (single young per nest; see  Supplemental Material Table S1 (169_rapt-55-02-09_s02.pdf) for data set 2 estimates). n, sample size; A, number of alleles; AR, allelic richness (95% CI); Ho, observed heterozygosity (± SE); He, expected heterozygosity (± SE); Fis, inbreeding coefficient (95% CI); r, mean pairwise relatedness.

img-z7-2_169.gif

Population Genetic Differentiation. A PCoA showed that the Aplomado Falcon individuals formed a single cluster with no clear demarcation of populations (Fig. 2). Axes 1, 2, and 3 explained 13.63%, 10.27%, and 8.30% of the variation, respectively, or 32.20% of the total. Similarly, no genetic differentiation was observed among founders and temporal periods within the South Texas population based on the analysis STRUCTURE. Although K = 9 had the highest mean probability and ΔK ( Supplemental Material 2, Table S2 (169_rapt-55-02-09_s02.pdf)), all the plots for K = 2 to 10 showed consistent admixture with no clustering of individuals by location or year, and therefore failed to indicate any evidence of population genetic structure among time periods or management areas ( Supplemental Material 2, Fig. S2 (169_rapt-55-02-09_s02.pdf)). Estimates of population differentiation using FST and Dest provided similar results indicating no consistent population structure among founders and each of the temporally sampled management areas after adjusting the significance level due to multiple comparisons ( Supplemental Material 2, Table S3 (169_rapt-55-02-09_s02.pdf); Rice 1989).

Figure 2.

Principal coordinate analysis (PCoA) using one Aplomado Falcon individual per nest for each year sampled and breeding area (Laguna Atascosa National Wildlife Refuge, LANWR; Matagorda Island National Wildlife Refuge, MINWR). Axis labels represent the first two coordinate axes with their respective explained variation. See online version for interpretation of color symbols.

img-z7-6_169.jpg

Effective Breeding (Nb) and Population (Ne) Size. Effective number of breeders (Nb) varied among sampled cohorts and populations, with estimates for South Texas ranging from 22 in 2013 to 39 in 2016 with overlapping 95% CIs among all sampled years (Table 3). The number of breeders estimated for each sampled cohort using the LD method (Nb(LD-Adj2)) ranged from 18 in 2014 to 61 in 2012 with undefined upper 95% CIs (or identified as infinity), which limited their use for comparative purposes (Table 3).

Adjusted estimates of the effective population size (Ne(LD-Adj2)) per generation varied among the sampled cohorts and ranged from 13 in 2014 to 44 in 2012, with all values smaller than their corresponding Nb(LD-Adj2) values for each cohort including three of their seven Nb(sib) estimates (Table 3). Using the pooled data set based on a single random young per nest and reproductive cohort for South Texas, Ne(LD-pool) was 21.0 (95% CI=10.6–40.3) and was comparable to the harmonic mean of Ne(LD-Adj2) obtained for all cohorts (harmonic mean=24.8). In contrast, the estimate of Ne for the founder population was large (Ne(LD-pool) = 8901.9) with high sampling variance as shown by the extremely wide 95% CI (31.4–∞).

Discussion

Population genetic analyses of the South Texas Aplomado Falcon population between 2004–2005 and 2012–2016 indicate no significant change in genetic diversity or effective breeding size (Nb) over the sampled time period. In fact, the diversity measures were similar to estimates obtained from a subset of individuals (n = 11) from the founding population used in the captive propagation and release program. Estimates of Nb and Ne for each reproductive cohort for the South Texas population were similar or slightly higher than the actual number of wild-caught Aplomado Falcons (n = 25) used to create the captive population in the early 1990s. These results indicate that the wild population (at least through 2016) appeared stable and comparable in genetic diversity and number since the time of their reintroduction. Further, based on annual nest survey results, there was no indication that a consistent decline in productivity has occurred (P. Juergens and B. Mutch, unpubl. data).

Between 2008 and 2013, the South Texas population averaged 37 occupied territories per year (range: 30–41) and 31 nesting territories (range: 28–34; Hunt et al. 2013). Similar levels were observed in 2014 through the 2017 breeding season (P. Juergens and B. Mutch, unpubl. data), but occupancy levels were lower in 2018, likely due to habitat modification and nest site damage following Hurricane Harvey in August 2017. Survey efforts during 2018 revealed ten territorial pairs of Aplomado Falcons had been lost to Hurricane Harvey: five pairs on Matagorda Island National Wildlife Refuge and five pairs on San Jose Island. The single pair occupying Mustang Island State Park remained. The LANWR population was not affected by Hurricane Harvey but occupancy had declined from 21 pairs observed in 2017. Recent surveys in South Texas indicated that numbers have rebounded, with 28 Aplomado Falcon pairs with established territories (LANWR: 19; MINWR: 9) and 45 nestlings observed during the 2020 breeding season (P. Juergens and B. Mutch, unpubl. data).

Each of Nb(sib) and Nb(LD-Adj2) estimates were similar to twice the number of nests sampled for each of the reproductive cohorts and also similar to the harmonic mean estimate of Ne(LD-Adj2) for the South Texas Aplomado Falcon population (Table 3). These differences suggest few additional nests beyond those that were sampled for this study contributed to the breeding population for each cohort. Previous research has shown that artificial nest sites, which constituted the majority of nests sampled for this study, are essential for increasing hatch-year survival and productivity of the current South Texas population (Hunt et al. 2013, McClure et al. 2017a). These results further confirm that the breeding population relies heavily on the artificial structures for nest sites because estimates of Nb tend to exceed the population's Ne.

The accuracy of the sibling assignment method for estimating Nb is maximized when sample sizes are near to or greater than the true Nb (Wang 2009, 2016; Ackerman et al. 2017). In this case, we feel that we have achieved that level of confidence because only a minimal increase in estimates of Nb(sib) was observed when consecutive cohorts were included in the analysis with increasing sample size, i.e., 2004–2005 (n = 73), Nb(sib) = 42 (95% CI = 30–69); 2012–2014 (n = 86), Nb(sib) = 52 (95% CI = 36–80); 2014–2016 (n = 111), Nb(sib) = 53 (95% CI = 38–79). The sibship assignment method estimates Nb of the parent generation that produced the sampled cohort, and thus, we expect an increase in the estimate as new full- and half-sibling groups are included in the analysis with consecutive cohorts. The magnitude by which Nb(sib) increases will depend on the number of adult pairs that breed over consecutive years. These results indicate the need for future efforts to increase the number of artificial nest platforms in the study population (see also McClure et al. 2017a), thereby providing an opportunity to increase the number of breeding pairs and increase overall productivity of the recovering Aplomado Falcon population in South Texas.

Although we documented at least three cases where nestlings were identified as half-siblings based on pairwise estimates of rw ≈ 0.25 and diploid allelic inheritance patterns among nestlings (max number of alleles per family group=4), we do not know if the pattern was due to extra-pair paternity (EPP) or intraspecific brood parasitism. An additional four nests may also include half-siblings based on the upper 95% CI estimate of rw; however, we failed to reject a full-sibling relationship for the nestlings in those nests based on allelic inheritance patterns either because only two young were sampled at the nest or because insufficient allelic variability existed in the data set. Not having the genotypes of parents at each nest decreased our ability to identify extrapair young (EPY); this is common among species with small clutch sizes such as the Aplomado Falcon (three-egg clutches are typical; Keddy-Hector et al. 2020).

The rate of EPY observed in the South Texas population was low at only 1–3% of young and 3–6% of nests (n = 3 or 7 of 267 sampled nestlings and 3 or 7 of 108 nests depending on the two estimates of EPY in this study). Nonetheless, additional research is warranted to determine if similar patterns exist in other Aplomado Falcon populations elsewhere in their distribution or if the phenomenon is unique to the South Texas reintroduced population. Although the occurrence of EPY in the population can influence genetic variability within a brood and potentially increase reproductive success for a male that breeds with multiple females, indirect benefits, especially to the female, are less understood and can vary depending on the local environment (Brouwer and Griffith 2019). The incidence of EPY in some species is positively correlated with the relatedness of the social partners at the nest (Arct et al. 2015). Although we do not have genotype data of the adults at each nest to investigate this further, the overall frequency of EPY as reported in this study is low and unlikely to have a major effect at the population level from a management perspective.

Given the life history constraints of the Aplomado Falcon (i.e., long generation time and small brood size) and recency of captive propagation and release, continued efforts should be made to collect samples every year to continue monitoring genetic diversity levels. The current data set based on microsatellite genotypes was insufficient to completely assess connectivity between management areas within the South Texas population by using relatedness to assess dispersal patterns. If possible, adult birds should be identified within respective breeding territories based on their banding records, and researchers should also generate single nucleotide polymorphism (SNP) data using next generation sequencing methods to obtain higher precision in estimates of rw (Kopps et al. 2015, Ackerman et al. 2017, Wang 2017). We were limited in our ability based on the current data set to distinguish between familial relationships such as half-siblings, first cousins, and grandparent-grandnestling, and additional genetic markers (e.g., loci and SNPs) should allow us to identify potential sires of young that were the product of EPP (e.g., Weinman et al. 2015) and further investigate turnover of breeders at nest sites.

Supplemental Material (available online).

 Supplemental Material 1 (169_rapt-55-02-09_s01.pdf). Summary of founding Northern Aplomado Falcon population used for propagation and release of captive-hatched individuals.  Supplemental Material 2. Table S1 (169_rapt-55-02-09_s02.pdf): Microsatellite DNA (10 loci) diversity estimates for each Northern Aplomado Falcon population temporal period.  Table S2 (169_rapt-55-02-09_s02.pdf): Mean LnP(K) for K = 1 to 10 using dataset 1 (single young per nest) with program STRUCTURE.  Table S3 (169_rapt-55-02-09_s02.pdf): Pairwise estimates of a) FST and b) Dest among Northern Aplomado Falcon founders and the South Texas management areas (LANWR and MINWR) sampled in 2004–2005 and 2012–2016 using dataset 1 (single young per nest).  Figure S1 (169_rapt-55-02-09_s02.pdf): Box plots comparing the relatedness estimates of four of seven different estimators for simulated individuals of known relatedness using the program Related (Pew et al. 2015).  Figure S2 (169_rapt-55-02-09_s02.pdf): Assignment of individual Aplomado Falcons to a defined cluster using Structure for K = 2 to 4.

Acknowledgments

Funding was provided by The Peregrine Fund and the University of North Texas. We thank Dean Keddy-Hector and Alberto Macías-Duarte for providing information concerning Northern Aplomado Falcons in Mexico and their current distribution. Rick Watson provided helpful comments concerning an earlier version of this report, and Brian Rolek provided demographic parameters used in the analysis. Sarah Schulwitz sampled muscle tissues from frozen carcasses representing a subset of the founder captive population curated at The World Center of Birds of Prey located in Boise, Idaho.

Literature Cited

1.

Ackerman, M. S., P. Johri, K. Spitze, S. Xu, T. G. Doak, K. Young, and M. Lynch (2017a). Estimating seven coefficients of pairwise relatedness using population genomic data. Genetics 206:105–118. Google Scholar

2.

Ackerman, M. W., B. K. Hand, R. K. Waples, G. Luikart, R. S. Waples, C. A. Steele, B. A. Garner, J. McCane, and M. R. Campbell (2017b). Effective number of breeders from sibship reconstruction: Empirical evaluations using hatchery steelhead. Evolutionary Applications 10:146–160. Google Scholar

3.

Allendorf, F. W., G. H. Luikart, and S. N. Aitken (2012). Conservation and the Genetics of Populations, Second Ed. Wiley-Blackwell Publishing, Hoboken, NJ, USA. Google Scholar

4.

Arct, A., S. M. Drobniak, and M. Cichoń (2015). Genetic similarity between mates predicts extrapair paternity – A meta-analysis of bird studies. Behavioral Ecology 26:959–968. Google Scholar

5.

Beasley, R. R., S. L. Lance, K. L. Jones, R. B. Berry, and J. A. Johnson (2014). Development of polymorphic microsatellite markers for the Orange-breasted Falcon (Falco deiroleucus). Conservation Genetic Resources 6:743–745. Google Scholar

6.

Blouin, M. S. (2003). DNA-based methods for pedigree reconstruction and kinship analysis in natural populations. Trends in Ecology & Evolution 18:503–511. Google Scholar

7.

Brouwer, L., and S. C. Griffith (2019). Extra-pair paternity in birds. Molecular Ecology 28:4864–4882. Google Scholar

8.

Cade, T. J., J. P. Jenny, and B. J. Walton (1991). Efforts to restore the Northern Aplomado Falcon Falco femoralis septentrionalis by captive breeding and reintroduction. Dodo, Journal of the Jersey Wildlife Preservation Trust 27:71–81. Google Scholar

9.

Charlesworth, B. (2009). Effective population size and patterns of molecular evolution and variation. Nature Reviews Genetics 10:195–205. Google Scholar

10.

Csillery, K., T. Johnson, D. Beraldi, T. Clutton-Brock, D. Coltman, B. Hansson, G. Spong, and J. M. Pemberton (2006). Performance of marker-based relatedness estimators in natural populations of outbred vertebrates. Genetics 173:2091–2101. Google Scholar

11.

de Jager, D., P. Swarts, C. Harper, and P. Bloomer (2017). Friends and family: A software program for identification of unrelated individuals from molecular marker data. Molecular Ecology Resources 17:e225–e233. Google Scholar

12.

Do, C., R. S. Waples, D. Peel, G. M. MacBeth, B. J. Tillett, and J. R. Ovenden (2014). NeEstimator v2: Reimplementation of software for the estimation of contemporary effective population size (Ne) from genetic data. Molecular Ecology Resources 14:209–214. Google Scholar

13.

Earl, D. A., and B. M. vonHoldt (2012). STRUCTURE HARVESTER: a website and program for visualizing STRUCTURE output and implementing the Evanno method. Conservation Genetics Resources 4:359–361. Google Scholar

14.

Evanno, G., S. Regnaut, and J. Goudet (2005). Detecting the number of clusters of individuals using the software STRUCTURE: A simulation study. Molecular Ecology 14:2611–2620. Google Scholar

15.

Ewen, J. G., and D. P. Armstrong (2007). Strategic monitoring of reintroductions in ecological restoration programmes. Ecoscience 14:401–409. Google Scholar

16.

Funk, W. C., B. R. Forester, S. J. Converse, C. Darst, and S. Morey (2019). Improving conservation policy with genomics: A guide to integrating adaptive potential in the U.S. Endangered Species Act decisions for conservation practitioners and genetics. Conservation Genetics 20:115–134. Google Scholar

17.

Hare, M. P., L. Nunney, M. K. Schwartz, D. E. Ruzzante, M. Burford, R. S. Waples, K. Ruegg, and F. Palstra (2011). Understanding and estimating effective population size for practical applications in marine species management. Conservation Biology 25:438–449. Google Scholar

18.

Harrisson, K. A., A. Pavlova, M. Telonis-Scott, and P. Sunnucks (2014). Using genomics to characterize evolutionary potential for conservation of wild populations. Evolutionary Applications 7:1008–1025. Google Scholar

19.

Hill, W. G. (1981). Estimation of effective population size from data on linkage disequilibrium. Genetical Research 38:209–216. Google Scholar

20.

Hunt, W. G., J. L. Brown, T. J. Cade, J. Coffman, M. Curti, E. Gott, W. Heinrich, J. P. Jenny, P. Juergens, A. Macias-Duarte, A. B. Montoya, et al. (2013). Restoring Aplomado Falcons to the United States. Journal of Raptor Research 47:335–351. Google Scholar

21.

Jakobsson, M., and N. A. Rosenberg (2007). CLUMPP: A cluster matching and permutation program for dealing with label switching and multimodality in analysis of population structure. Bioinformatics 23:1801–1806. Google Scholar

22.

Jamieson, I. G., and F. W. Allendorf (2012). How does the 50/500 rule apply to MVPs. Trends in Ecology and Evolution 27:578–584. Google Scholar

23.

Jenny, J. P., W. Heinrich, A. B. Montoya, B. Mutch, C. Sandfort, and W. G. Hunt (2004). Progress in restoring the Aplomado Falcon to southern Texas. Wildlife Society Bulletin 32:276–285. Google Scholar

24.

Jones, O. R., and J. Wang (2010). COLONY: A program for parentage and sibship inference from multilocus genotype data. Molecular Ecology Resources 10:551–555. Google Scholar

25.

Jost, L. (2008). GST and its relatives do not measure differentiation. Molecular Ecology 17:4015–4026. Google Scholar

26.

Keddy-Hector, D. P. (2019). The history of Aplomado Falcon Falco femoralis subspecies diagnoses. Bulletin of the British Ornithologists' Club 139:111–126. Google Scholar

27.

Keddy-Hector, D. P., P. Pyle, and M. A. Patten (2020). Aplomado Falcon (Falco femoralis), version 1.0. InBirds of the World ( P. G. Rodewald, Editor). Cornell Lab of Ornithology, Ithaca, NY, USA.  https://doi.org/10.2173/bow.aplfal.01Google Scholar

28.

Keenan, K., P. McGinnity, T. F. Cross, W. W. Crozier, and P. A. Prodöhl (2013). diveRsity: An R package for the estimation of population genetics parameters and their associated errors. Methods in Ecology and Evolution 4:782–788. Google Scholar

29.

Kopps, A. M., J. Kang, W. B. Sherwin, and P. J. Palsbøkk (2015). How well do molecular and pedigree relatedness correspond, in populations with diverse mating systems, and various types and quantities of molecular and demographic data. G3 (Bethesda) 5:1815–1826. Google Scholar

30.

Lai, Y.-T., C. K. L. Yeung, K. E. Omland, E.-L. Pang, Y. Hao, B.-Y. Liao, H.-F. Cao, B.-W. Zhang, C.-F. Yeh, C.-M. Hung, H.-Y. Hung, et al. (2019). Standing genetic variation as the predominant source for adaptation of a songbird. Proceedings of the National Academy of Sciences USA 116:2152–2157. Google Scholar

31.

Leberg, P. L. (2002). Estimating allelic richness: Effects of sample size and bottlenecks. Molecular Ecology 11:2445–2449. Google Scholar

32.

Lewis, P. O., and D. Zaykin (2001). Genetic data analysis: Computer program for the analysis of allelic data. Version 1.0 (d16c).  http://lewis.eeb.uconn.edu/lewishome/software.htmlGoogle Scholar

33.

Luikart, G., N. Ryman, D. A. Tallmon, M. K. Schwartz, and F. W. Allendorf (2010). Estimation of census and effective population sizes: The increasing usefulness of DNA-based approaches. Conservation Genetics 11:355–373. Google Scholar

34.

McClure, C. J. W., B. P. Pauli, B. Mutch, and P. Juergens (2017a). Assessing the importance of artificial nest-sites in the population dynamics of endangered Northern Aplomado Falcons Falco femoralis septentrionalis in south Texas using stochastic simulation models. Ibis 159:14–25. Google Scholar

35.

McClure, C. J. W., B. W. Rolek, T. I. Hayes, C. D. Hayes, M. Curti, and D. L. Anderson (2017b). Successful enhancement of Ridgway's Hawk populations through recruitment of translocated birds. The Condor 119:855–864. Google Scholar

36.

Meirmans, P. G., and P. W. Hedrick (2011). Assessing population structure: FST and related measures. Molecular Ecology Resources 11:5–18. Google Scholar

37.

Peakall, R. O. D., and P. E. Smouse (2012). GenAlEx 6.5: Genetic analysis in Excel. Population genetic software for teaching and research – An update. Bioinformatics 28:2537–2539. Google Scholar

38.

Pew, J., P. H. Muir, J. Wang, and T. R. Frasier (2015). related: An R package for analyzing pairwise relatedness from codominant molecular markers. Molecular Ecology Resources 15:557–561. Google Scholar

39.

Pritchard, J. K., M. Stephens, and P. Donnelly (2000). Inference of population structure using multilocus genotype data. Genetics 155:945–959. Google Scholar

40.

R Core Team (2017). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria.  https://www.R-project.orgGoogle Scholar

41.

Ramasamy, R. K., S. Ramasamy, B. B. Bindroo, and V. G. Naik (2014). STRUCTURE PLOT: A program for drawing elegant STRUCTURE bar plots in user friendly interface. Springerplus 3:431. Google Scholar

42.

Rice, W. R. (1989). Analyzing tables of statistical tests. Evolution 43:223–225. Google Scholar

43.

Rodriguez-Ramilo, S. T., and J. Wang (2012). The effect of close relatives on unsupervised Bayesian clustering algorithms in population genetic structure analysis. Molecular Ecology Resources 12:873–884. Google Scholar

44.

Schwartz, M. K., G. Luikart, and R. Waples (2007). Genetic monitoring as a promising tool for conservation and management. Trends in Ecology and Evolution 22:24–33. Google Scholar

45.

Seddon, P. J., D. P. Armstrong, and R. F. Maloney (2007). Developing the science of reintroduction biology. Conservation Biology 21:303–312. Google Scholar

46.

Smouse, P. E., and R. Peakall (1999). Spatial autocorrelation analysis of individual multiallele and multilocus genetic structure. Heredity 82:561–573. Google Scholar

47.

Sweikert, L., and M. Phillips (2015). The effect of supplemental feeding on the known survival of reintroduced Aplomado Falcons: Implications for recovery. Journal of Raptor Research 49:389–399. Google Scholar

48.

Taylor, H. R. (2015). The use and abuse of genetic marker-based estimates of relatedness and inbreeding. Ecology and Evolution 5:3140–3150. Google Scholar

49.

US Fish and Wildlife Service (2014). Northern Aplomado Falcon (Falco femoralis septentrionalis) 5-Year Review: Summary and Evaluation. USDI Fish and Wildlife Service, New Mexico Ecological Services Field Office, Albuquerque, NM, USA. Google Scholar

50.

Van de Casteele, T., P. Galbusera, and E. Matthysen (2001). A comparison of microsatellite-based pairwise relatedness estimators. Molecular Ecology 10:1539–1549. Google Scholar

51.

Wang, J. (2002). An estimator for pairwise relatedness using molecular markers. Genetics 160:1203–1215. Google Scholar

52.

Wang, J. (2009). A new method for estimating population sizes from a single sample of multilocus genotypes. Molecular Ecology 18:2148–2164. Google Scholar

53.

Wang, J. (2011). COANCESTRY: A program for simulating, estimating and analyzing relatedness and inbreeding coefficients. Molecular Ecology Resources 11:141–145. Google Scholar

54.

Wang, J. (2016). A comparison of single-sample estimators of effective population sizes from genetic marker data. Molecular Ecology 25:4692–4711. Google Scholar

55.

Wang, J. (2017). Pedigrees or markers: which are better in estimating relatedness and inbreeding coefficient? Theoretical Population Biology 107:4–13. Google Scholar

56.

Waples, R. S. (2006). A bias correction for estimates of effective population size based on linkage disequilibrium at unlinked gene loci*. Conservation Genetics 7:167–184. Google Scholar

57.

Waples, R. S., and E. C. Anderson (2017). Purging putative siblings from population genetic data sets: A cautionary tale. Molecular Ecology 26:1211–1224. Google Scholar

58.

Waples, R. S., T. Antao, and G. Luikart (2014). Effects of overlapping generations on linkage disequilibrium estimates of effective population size. Genetics 197:769–780. Google Scholar

59.

Waples, R. S., and C. H. I. Do (2008). LDNE: A program for estimating effective population size from data on linkage disequilibrium. Molecular Ecology Resources 8:753–756. Google Scholar

60.

Waples, R. S., G. Luikart, J. R. Faulkner, and D. A. Tallmon (2013). Simple life-history traits explain key effective population size ratios across diverse taxa. Proceedings of the Royal Society B 280:20131339. Google Scholar

61.

Weinman, L. R., J. W. Solomon, and D. R. Rubenstein (2015). A comparison of single nucleotide polymorphism and microsatellite markers for analysis of parentage and kinship in a cooperatively breeding bird. Molecular Ecology Resources 7:1008–1025. Google Scholar

62.

Willoughby, J. R., M. Sundaram, B. K. Wijayawardena, S. J. A. Kimble, Y. Ji, N. B. Fernandez, J. D. Antonides, M. C. Lamb, N. J. Marra, and J. A. DeWoody (2015). The reduction of genetic diversity in threatened vertebrates and new recommendations regarding IUCN conservation rankings. Biological Conservation 191:495–503. Google Scholar

63.

Wright, S. (1931). Evolution in Mendelian populations. Genetics 16:97–159. Google Scholar

64.

Wright, S. (1938). Size of population and breeding structure in relation to evolution. Science 87:430–431. Google Scholar
© 2021 The Raptor Research Foundation, Inc.
Jeff A. Johnson, Alexandra Stock, Paul Juergens, Brian Mutch, and Christopher J.W. McClure "Temporal Genetic Diversity and Effective Population Size of the Reintroduced Aplomado Falcon (Falco femoralis) Population in Coastal South Texas," Journal of Raptor Research 55(2), 169-180, (8 June 2021). https://doi.org/10.3356/0892-1016-55.2.169
Received: 28 July 2020; Accepted: 4 January 2021; Published: 8 June 2021
KEYWORDS
Aplomado Falcon
effective population size
extra-pair young
Falco femoralis
genetic diversity
microsatellite DNA
reintroduction
Back to Top