Assessment of costs of reproduction in a pelagic seabird using multistate mark-recapture models

We used a long‐term population band‐resight survey database, a parallel reproduction database, and multistate mark–recapture analysis to assess the costs of reproduction, a keystone concept of life‐history evolution, in Nazca boobies (Sula granti) from Punta Cevallos, Isla Española, Galápagos, Ecuador. We used eight years of resight and breeding data to compare models that included sex‐ and state‐specific survival probabilities and probabilities of transition between reproductive states using multistate mark–recapture models. Models that included state‐specific effects were compared with models lacking such effects to evaluate costs of reproduction. The top model, optimizing the trade‐off of model simplicity and fit to the data using the Akaike Information Criterion (AIC), showed evidence of a temporally varying survival cost of reproduction: nonbreeders showed higher annual survival than breeders did in some years. Because increasing investment among breeders showed no negative association with survival and subsequent breeding success, this evidence indicates a cost to both males and females of initiating, but not of continuing, a reproductive attempt. In some cases, breeders reaching the highest reproductive state (fledging an offspring) showed higher survival or subsequent breeding success than did failed breeders, consistent with differences in overall quality that promote both survival and reproduction. Although a male‐biased adult sex ratio was observed in this population of Nazca boobies, models of state‐ and sex‐specific survival and transition probabilities were not supported, indicating that males and females do not incur different costs of reproduction, and that the observed sex ratio bias is not due to sex‐specific adult mortality.     

Effect of flowering on vegetative growth and further reproduction in

Flowering intensity and plant size were monitored in 155 Festuca novae-zelandiae individuals over four years to determine if trade-offs exist between inflorescence production and vegetative growth, and between inflorescence production in different years. Less than half of the population flowered in any one year, 36% of individuals did not flower at all, and only 17% flowered in all four years of the study. Mean number of inflorescences per individual per year varied from 1.54 to 5.53 (maximum = 85). No trade-offs were detected between flowering frequency and intensity; individuals that flowered more frequently also produced more inflorescences in each flowering episode. No trade-off was detected between current and future reproduction, rather flowering intensity was positively correlated between years. Growth, as measured by diameter increment, was positively related to flowering frequency and flowering intensity, both across all individuals studied and within 1m x 1m plots. The presence of a positive relationship between growth and reproduction within plots argues against meso-scale variability in environment factors being the cause of the results from analyses involving all individuals. Clearly reproduction in F. novae-zelandiae does not incur a marked cost in growth or future reproduction. The assumptions underlying theoretical expectations of such trade- offs may not be valid for long-lived clonal plants such as F. novae-zelandiae.     

Is analysing the nitrogen use at the plant canopy level a matter of choosing the right optimization criterion?

Herbivores can have strong deleterious effects on plant growth, reproduction, and even survival. Because these effects might be strongly interrelated, the direct consumptive effects of herbivores and a variety of indirect effects are difficult to untangle. Reductions in growth, for example, may strongly impact the flowering behaviour of plant species in the current season, but at the same time incur costs to survival, growth and reproduction in the next growing season(s). To get better insights in the effects of herbivory on the flowering behaviour of the long-lived polycarpic grassland herb Primula veris L., flowering patterns were monitored over ten consecutive years under two treatments (grazing and control mowing regimes). We tested the hypothesis that the size at flowering was affected by the presence of herbivores, and whether this translated into costs to future reproduction and survival. Overall, grazed plants were significantly smaller than control plants, and the size at which plants flowered was also significantly smaller when herbivores were present. The transition probability of flowering and of surviving into the next year was significantly smaller for all plants in the current year if they had been grazed than if they had been mown, indicating that herbivory incurred costs to both flowering and survival. Grazed plants also needed longer to start flowering, had fewer flowers and flowered less frequently, causing a significantly lower proportion of flowering adults in the population. These results suggest that the observed regression in plant size due to herbivory does not allow plants to capture enough resources to guarantee regular flowering in the longer run.     

Reproduction and development of the endangered Sedum integrifolium ssp. leedyi (Crassulaceae)

Information on reproduction and life history is important for the conservation of endangered plants. We investigated rates of flowering, seed set, and germination in populations of the endangered perennial plant Sedum integrifolium ssp. leedyi. Germination and flowering rates differed significantly among populations, but seed set rate did not. We assayed 26 plant clusters (81 stems) from four of the five known populations for evidence of clonal reproduction using 28 randomly amplified polymorphic DNA (RAPD) markers. Of the 81 stems, 75 had unique genotypes and three pairs had identical genotypes, suggesting that clonal reproduction is infrequent. Flowering, seed set, and germination rates were correlated with our estimates of ratios of effective to actual population sizes (Ne/N), but not with Ne. The single formally protected population may be experiencing inbreeding depression. We grew plants from seed to maturity in a greenhouse, with a germination rate of 77% and survival of 98% of the germinants at 6 mo, suggesting that this will be a viable means of ex situ propagation. Plants flowered 4-6 mo after germination and produced mature fruits 1-2 mo later, suggesting that they have the potential to sexually reproduce in their first or second season of growth.     

Apparent survival and return rate of breeders in the southern temperate White‐rumped Swallow Tachycineta leucorrhoa

Life‐histories and demographic parameters of southern temperate bird species have been little studied. We estimated return rates between years and sexes, and adult apparent survival and recapture probabilities with mark–recapture data on White‐rumped Swallows and found a lower return rate of unsuccessful females. There was little support for influences of sex or year on survival rates. The estimates were equivalent to the lowest value reported for a northern congener, in contrast to the prediction of geographical variation under life‐history theory.     

Evaluating environmental,demographic and genetic effects on population‐level survival in an island endemic

The population dynamics of island species are considered particularly sensitive to variation in environmental, demographic and/or genetic processes. However, few studies have attempted to evaluate the relative importance of these processes for key vital rates in island endemics. We integrated the results of long‐term capture–mark–recapture analysis, prey surveys, habitat quality assessments and molecular analysis to determine the causes of variation in the survival rates of Komodo dragons Varanus komodoensis at 10 sites on four islands in Komodo National Park, Indonesia. Using open population capture–mark–recapture methods, we ranked competing models that considered environmental, ecological, genetic and demographic effects on site‐specific Komodo dragon survival rates. Site‐specific survival rates ranged from 0.49 (95% CI: 0.33–0.68) to 0.92 (0.79–0.97) in the 10 study sites. The three highest‐ranked models (i.e. ΔQAIC_c < 2) explained ～70% of variation in Komodo dragon survival rates and identified interactions between inbreeding coefficients, prey biomass density and habitat quality as important explanatory variables. There was evidence of additive effects from ecological and genetic (e.g. inbreeding) processes affecting Komodo dragon survival rates. Our results indicate that maintaining high ungulate prey biomass and habitat quality would enhance the persistence of Komodo dragon populations. Assisted gene flow may also increase the genetic and demographic viability of the smaller Komodo dragon populations.     

THE FLORAL ECOLOGY OF PLATANTHERA BLEPHARIGLOTTIS

The floral ecology of a central Maine population of Platanthera blephariglottis was studied during the 1979–1981 flowering seasons. Although this species possesses characteristics typical of moth-pollinated plants, only diurnal pollinators were documented, primarily Hesperiidae, Pieridae, and Apidae. Pollinators were scarce throughout the study, yet bagging experiments indicated vectors were necessary for successful capsule set. The average capsule set per plant was lowest (32.2%) in 1979 when 362 plants bloomed and highest in 1980 (80.2%) when 202 plants flowered. Capsule set was found to be limited by pollinator activity. The adaptations of P. blephariglottis, including sequential flowering, length of receptivity, nectar production and inflorescence size, enable consistently successful reproduction regardless of microhabitat. This species is well-adapted to reproduce in environments experiencing periods of unfavorable weather and low pollinator abundance.     

Evolutionary demography of iteroparous plants: incorporating non-lethal costs of reproduction into integral projection models

Understanding the selective forces that shape reproductive strategies is a central goal of evolutionary ecology. Selection on the timing of reproduction is well studied in semelparous organisms because the cost of reproduction (death) can be easily incorporated into demographic models. Iteroparous organisms also exhibit delayed reproduction and experience reproductive costs, although these are not necessarily lethal. How non-lethal costs shape iteroparous life histories remains unresolved. We analysed long-term demographic data for the iteroparous orchid Orchis purpurea from two habitat types (light and shade). In both the habitats, flowering plants had lower growth rates and this cost was greater for smaller plants. We detected an additional growth cost of fruit production in the light habitat. We incorporated these non-lethal costs into integral projection models to identify the flowering size that maximizes fitness. In both habitats, observed flowering sizes were well predicted by the models. We also estimated optimal parameters for size-dependent flowering effort, but found a strong mismatch with the observed flower production. Our study highlights the role of context-dependent non-lethal reproductive costs as selective forces in the evolution of iteroparous life histories, and provides a novel and broadly applicable approach to studying the evolutionary demography of iteroparous organisms.     

COST OF REPRODUCTION IN POLEMONIUM VISCOSUM: PHENOTYPIC AND GENETIC APPROACHES

I measured the effect of early reproduction on subsequent growth and survival in the alpine perennial wildflower, Polemonium viscosum. Measurements were made over 4 yr on 34 maternal sibships under natural conditions. A significant phenotypic cost of early reproduction characterized the study population. Plants that flowered after only one year's growth had twice as many leaves and 25% more shoots than nonflowering individuals of equal age. However, early flowering decreased leaf number by 18% in the subsequent year and survivorship by 20% after two years relative to changes in leaf number and survival of nonflowering plants. For such trade-offs to shape the further evolution of reproductive schedules, flowering probability and those age-specific components of plant size that represent the energetic currency for reproductive costs must be heritable. Although families showed significant heterogeneity in the probability of early flowering, most (62%) entirely failed to flower. Moreover, phenotypic variation in vegetative size components at ages 1 and 2 had little genetic basis. Only at ages 3 and 4, after vegetative and demographic costs of early reproduction had been incurred, did vegetative size components (leaf length and number, and shoot number) vary significantly among families. Results of this study provide little evidence of a genetically based trade-off between early reproduction and subsequent survival in P. viscosum.     

Demographic estimation methods for plants with unobservable life-states

Demographic estimation of vital parameters in plants with an unobservable dormant state is complicated, because time of death is not known. Conventional methods assume that death occurs at a particular time after a plant has last been seen aboveground but the consequences of assuming a particular duration of dormancy have never been tested. Capture–recapture methods do not make assumptions about time of death; however, problems with parameter estimability have not yet been resolved. To date, a critical comparative assessment of these methods is lacking. We analysed data from a 10 year study of Cleistes bifaria, a terrestrial orchid with frequent dormancy, and compared demographic estimates obtained by five varieties of the conventional methods, and two capture–recapture methods. All conventional methods produced spurious unity survival estimates for some years or for some states, and estimates of demographic rates sensitive to the time of death assumption. In contrast, capture–recapture methods are more parsimonious in terms of assumptions, are based on well founded theory and did not produce spurious estimates. In Cleistes, dormant episodes lasted for 1–4 years (mean 1.4, SD 0.74). The capture–recapture models estimated ramet survival rate at 0.86 (SE～0.01), ranging from 0.77–0.94 (SEs≤0.1) in any one year. The average fraction dormant was estimated at 30% (SE 1.5), ranging 16–47% (SEs≤5.1) in any one year. Multistate capture–recapture models showed that survival rates were positively related to precipitation in the current year, but transition rates were more strongly related to precipitation in the previous than in the current year, with more ramets going dormant following dry years. Not all capture–recapture models of interest have estimable parameters; for instance, without excavating plants in years when they do not appear aboveground, it is not possible to obtain independent time‐specific survival estimates for dormant plants. We introduce rigorous computer algebra methods to identify the parameters that are estimable in principle. As life‐states are a prominent feature in plant life cycles, multistate capture–recapture models are a natural framework for analysing population dynamics of plants with dormancy.     

An integrated population model for a long‐lived ungulate: more efficient data use with Bayesian methods

We develop an integrated population model for Svalbard reindeer Rangifer tarandus platyrhynchus, and demonstrate how this type of model can be used to extract more information from the data and separate different sources of variability in population estimates. Our model combines individual mark–recapture data with population counts and harvesting data within a Bayesian model framework, and accounts for observation error, environmental and demographic stochasticity, and age structure. From this model we obtain annual estimates of age‐specific population size, survival and fecundity. The model provides estimates of age structure at a finer scale than that found in the census data, and enables us to estimate survival for the period before calves are first caught and marked, i.e. before they enter the individual mark–recapture data. The modeling framework provides an improved approach to studying age‐structured populations that are imperfectly censused and where the demography of only a sample of individuals is known. We use data from independent censuses of the same population to evaluate population estimates obtained from the model, and show that it is successful at correcting for different types of observation error. Based on our model results, we suggest that allocating resources to the collection of supplementary mark–recapture data could improve the reliability of population projections more than making regular population censuses as exhaustive as possible. Our work demonstrates how integrated Bayesian population modeling can be used to increase the amount of information extracted from collections of data, identifying and disentangling sources of variation in individual performance and population size. This represents an important step towards increasing the predictive ability of population growth models for long‐lived species experiencing changes in environmental conditions and harvesting regimes.     

Improved methods for estimating abundance and related demographic parameters from mark‐resight data

Over the past decade, there has been much methodological development for the estimation of abundance and related demographic parameters using mark‐resight data. Often viewed as a less‐invasive and less‐expensive alternative to conventional mark recapture, mark‐resight methods jointly model marked individual encounters and counts of unmarked individuals, and recent extensions accommodate common challenges associated with imperfect detection. When these challenges include both individual detection heterogeneity and an unknown marked sample size, we demonstrate several deficiencies associated with the most widely used mark‐resight models currently implemented in the popular capture‐recapture freeware Program MARK. We propose a composite likelihood solution based on a zero‐inflated Poisson log‐normal model and find the performance of this new estimator to be superior in terms of bias and confidence interval coverage. Under Pollock's robust design, we also extend the models to accommodate individual‐level random effects across sampling occasions as a potentially more realistic alternative to models that assume independence. As a motivating example, we revisit a previous analysis of mark‐resight data for the New Zealand Robin (Petroica australis) and compare inferences from the proposed estimators. For the all‐too‐common situation where encounter rates are low, individual detection heterogeneity is non‐negligible, and the number of marked individuals is unknown, we recommend practitioners use the zero‐inflated Poisson log‐normal mark‐resight estimator as now implemented in Program MARK.     

INVESTIGATING EVOLUTIONARY TRADE‐OFFS IN WILD POPULATIONS OF ATLANTIC SALMON (SALMO SALAR): INCORPORATING DETECTION PROBABILITIES AND INDIVIDUAL HETEROGENEITY

Evolutionary trade‐offs among demographic parameters are important determinants of life‐history evolution. Investigating such trade‐offs under natural conditions has been limited by inappropriate analytical methods that fail to address the bias in demographic estimates that can result when issues of detection (uncertain detection of individual) are ignored. We propose a new statistical approach to quantify evolutionary trade‐offs in wild populations. Our method is based on a state‐space modeling framework that focuses on both the demographic process of interest as well as the observation process. As a case study, we used individual mark–recapture data for stream‐dwelling Atlantic salmon juveniles in the Scorff River (Southern Brittany, France). In freshwater, juveniles face two life‐history choices: migration to the ocean and sexual maturation (for males). Trade‐offs may appear with these life‐history choices and survival, because all are energy dependent. We found a cost of reproduction on survival for fish staying in freshwater and a survival advantage associated with the “decision” to migrate. Our modeling framework opens up promising prospects for the study of evolutionary trade‐offs when some life‐history traits are not, or only partially, observable.     

The Demography of Miconia prasina (Melastomataceae) During Secondary Succession in Puerto Rico1

Miconia prasina (Melastomataceae) is an important colonizing species during early stages of secondary succession in Puerto Rico but its abundance declines with increasing stand age. We studied its demography for 2 yr (1995–1997) in three populations across a pasture to forest chronosequence (1‐, 12‐, and 25‐yr old) and in one population in a hurricane‐disturbed older forest (>60‐yr old). Vegetative growth was positive in the two young sites, but negative in the two older sites. The highest percentage of plants that flowered occurred in the two younger sites but highest seed production occurred in the 25‐yr‐old site. Although seed germination occurred in the two older sites, no seedlings established. Vegetative reproduction (root sprouts) was found in all sites, with most sprouts occurring in young sites. The 2 yr of census data were used to construct stage‐based transition matrices. In the two young sites, the average population growth rate was positive (λ > 1), while average population growth was negative in the two older sites (λ < 1). Elasticity, a measure of the effects of proportional changes in life‐history transitions to the long‐term population growth rate, varied across both stages and sites. Elasticity shifted from large plants in young sites to small plants in older sites. Across all sites, elasticities related to survival were more important than those of growth or vegetative reproduction. The demographic analysis supported the previous observations of changes in the density of M. prasina based on the chronosequence and also suggests that establishment via seeds must occur prior to abandonment in active pastures.     

Population biology of the rare military orchid (Orchis militaris L.) at an established site in Suffolk, England

In Britain, where it reaches the north-westerly limit of its European distribution, Orchis militaris L. is extremely rare. Well-established and persistent populations of O. militaris are known to exist at only two sites. The largest extant population of O. militaris occurs in a disused chalk pit in Suffolk. A preliminary demographic analysis of this population, covering the period 1975 to 1991, along with estimates of key life stage transition probabilities are presented here. From 1975 to 1986 the number of separate identifiable plants in the population declined substantially. Until 1986 recruitment of rosettes was poor. The largest cohort of new plants, recorded in 1976, was 35. Approximately 48% of new individuals recruited between 1976 and 1985 failed to flower. Of those flowering, approximately 55% flowered during their first year above ground. Of the original population recorded in 1975, 67.8% flowered at least once during the study. The reproductive performance, i.e. the frequency of flowering and the period between episodes of flowering, varied considerably between individuals. Some plants flowered every year while others only flowered once during the study. Few plants remained below ground for more than one year, while several apparently persisted below ground for more than 6 successive years. Although the number of plants that can be identified as separate individuals has declined, the total number of rosettes in the population has, from 1986, increased dramatically. Because of the dense clumping of these recruits it is not possible to determine whether they are derived from seed or vegetative propagation. When post 1986 recruitment is combined with the number of plants that established before 1986 and survive, the apparent number of plants present at the site has more than doubled between 1975 and1991.     

Selection on flowering time in Mediterranean high-mountain plants under global warming

Under climate warming, plants will undergo novel selective pressures to adjust reproductive timing. Adjustment between reproductive phenology and environment is expected to be higher in arctic and alpine habitats because the growing season is considerably short. As early- and late-flowering species reproduce under very different environmental conditions, selective pressures on flowering phenology and potential effects of climate change are likely to differ between them. However, there is no agreement on the magnitude of the benefits and costs of early- vs. late-flowering species under a global warming scenario. In spite of its relevance, phenotypic selection on flowering phenology has rarely been explored in alpine plants and never in Mediterranean high mountain species, where selective pressures are very different due to the summer drought imposed over the short growth season. We hypothesized that late-flowering plants in Mediterranean mountains should present stronger selective pressures towards early onset of reproduction than early-flowering species, because less water is available in the soil as growing season progresses. We performed selection analyses on flowering onset and duration in two high mountain species of contrasting phenology. Since phenotypic selection can be highly context-dependent, we studied several populations of each species for 2 years, covering their local altitudinal ranges and their different microhabitats. Surrogates of biotic selective agents, like fruitset for pollinators and flower and fruit loss for flower and seed predators, were included in the analysis. Differences between the early- and the late-flowering species were less than expected. A consistent negative correlational selection of flowering onset and duration was found affecting plant fitness, i.e., plants that bloomed earlier flowered for longer periods improving plant fitness. Nevertheless, the late-flowering species may experience higher risks under climate warming because in extremely warm and dry years the earlier season does not bring about a longer flowering duration due to summer drought.     

Demographic consequences of changing body size in a terrestrial salamander

Changes in climate can alter individual body size, and the resulting shifts in reproduction and survival are expected to impact population dynamics and viability. However, appropriate methods to account for size‐dependent demographic changes are needed, especially in understudied yet threatened groups such as amphibians. We investigated individual‐ and population‐level demographic effects of changes in body size for a terrestrial salamander using capture–mark–recapture data. For our analysis, we implemented an integral projection model parameterized with capture–recapture likelihood estimates from a Bayesian framework. Our study combines survival and growth data from a single dataset to quantify the influence of size on survival while including different sources of uncertainty around these parameters, demonstrating how selective forces can be studied in populations with limited data and incomplete recaptures. We found a strong dependency of the population growth rate on changes in individual size, mediated by potential changes in selection on mean body size and on maximum body size. Our approach of simultaneous parameter estimation can be extended across taxa to identify eco‐evolutionary mechanisms acting on size‐specific vital rates, and thus shaping population dynamics and viability.     

Phenological change in a spring ephemeral: implications for pollination and plant reproduction

Climate change has had numerous ecological effects, including species range shifts and altered phenology. Altering flowering phenology often affects plant reproduction, but the mechanisms behind these changes are not well‐understood. To investigate why altering flowering phenology affects plant reproduction, we manipulated flowering phenology of the spring herb Claytonia lanceolata (Portulacaceae) using two methods: in 2011–2013 by altering snow pack (snow‐removal vs. control treatments), and in 2013 by inducing flowering in a greenhouse before placing plants in experimental outdoor arrays (early, control, and late treatments). We measured flowering phenology, pollinator visitation, plant reproduction (fruit and seed set), and pollen limitation. Flowering occurred approx. 10 days earlier in snow‐removal than control plots during all years of snow manipulation. Pollinator visitation patterns and strength of pollen limitation varied with snow treatments, and among years. Plants in the snow removal treatment were more likely to experience frost damage, and frost‐damaged plants suffered low reproduction despite lack of pollen limitation. Plants in the snow removal treatment that escaped frost damage had higher pollinator visitation rates and reproduction than controls. The results of the array experiment supported the results of the snow manipulations. Plants in the early and late treatments suffered very low reproduction due either to severe frost damage (early treatment) or low pollinator visitation (late treatment) relative to control plants. Thus, plants face tradeoffs with advanced flowering time. While early‐flowering plants can reap the benefits of enhanced pollination services, they do so at the cost of increased susceptibility to frost damage that can overwhelm any benefit of flowering early. In contrast, delayed flowering results in dramatic reductions in plant reproduction through reduced pollination. Our results suggest that climate change may constrain the success of early‐flowering plants not through plant‐pollinator mismatch but through the direct impacts of extreme environmental conditions.     

Individual heterogeneity and capture–recapture models: what,why and how?

Variation between and within individuals in life history traits is ubiquitous in natural populations. When affecting fitness‐related traits such as survival or reproduction, individual heterogeneity plays a key role in population dynamics and life history evolution. However, it is only recently that properly accounting for individual heterogeneity when studying population dynamics of free‐ranging populations has been made possible through the development of appropriate statistical models. We aim here to review case studies of individual heterogeneity in the context of capture–recapture models for the estimation of population size and demographic parameters with imperfect detection. First, we define what individual heterogeneity means and clarify the terminology used in the literature. Second, we review the literature and illustrate why individual heterogeneity is used in capture–recapture studies by focusing on the detection of life‐history tradeoffs, including senescence. Third, we explain how to model individual heterogeneity in capture–recapture models and provide the code to fit these models ( https://github.com/oliviergimenez/indhet_in_CRmodels ). The distinction is made between situations in which heterogeneity is actually measured and situations in which part of the heterogeneity remains unobserved. Regarding the latter, we outline recent developments of random‐effect models and finite‐mixture models. Finally, we discuss several avenues for future research.