Strategies for fitting nonlinear ecological models in R, AD Model Builder, and BUGS

Plant traits poorly predict winner and loser shrub species in a warming tundra biome

Climate change is leading to species redistributions. In the tundra biome, shrubs are generally expanding, but not all tundra shrub species will benefit from warming. Winner and loser species, and the characteristics that may determine success or failure, have not yet been fully identified. Here, we investigate whether past abundance changes, current range sizes and projected range shifts derived from species distribution models are related to plant trait values and intraspecific trait variation. We combined 17,921 trait records with observed past and modelled future distributions from 62 tundra shrub species across three continents. We found that species with greater variation in seed mass and specific leaf area had larger projected range shifts, and projected winner species had greater seed mass values. However, trait values and variation were not consistently related to current and projected ranges, nor to past abundance change. Overall, our findings indicate that abundance change and range shifts will not lead to directional modifications in shrub trait composition, since winner and loser species share relatively similar trait spaces.

Non-linear models of species' responses to environmental and spatial gradients

Species' responses to broad-scale environmental or spatial gradients are typically unimodal. Current models of species' responses along gradients tend to be overly simplistic (e.g., linear, quadratic or Gaussian GLMs), or are suitably flexible (e.g., splines, GAMs) but lack direct ecologically interpretable parameters. We describe a parametric framework for species-environment non-linear modelling ('senlm'). The framework has two components: (i) a non-linear parametric mathematical function to model the mean species response along a gradient that allows asymmetry, flattening/peakedness or bimodality; and (ii) a statistical error distribution tailored for ecological data types, allowing intrinsic mean-variance relationships and zero-inflation. We demonstrate the utility of this model framework, highlighting the flexibility of a range of possible mean functions and a broad range of potential error distributions, in analyses of fish species' abundances along a depth gradient, and how they change over time and at different latitudes.

Preliminary evidence for one-trial social learning of vervet monkey alarm calling

How do non-human primates learn to use their alarm calls? Social learning is a promising candidate, but its role in the acquisition of meaning and call usage has not been studied systematically, neither during ontogeny nor in adulthood. To investigate the role of social learning in alarm call comprehension and use, we exposed groups of wild vervet monkeys to two unfamiliar animal models in the presence or absence of conspecific alarm calls. To assess the learning outcome of these experiences, we then presented the models for a second time to the same monkeys, but now without additional alarm call information. In subjects previously exposed in conjunction with alarm calls, we found heightened predator inspection compared to control subjects exposed without alarm calls, indicating one-trial social learning of 'meaning'. Moreover, some juveniles (but not adults) produced the same alarm calls they heard during the initial exposure whereas the authenticity of the models had an additional effect. Our experiment provides preliminary evidence that, in non-human primates, call meaning can be acquired by one-trail social learning but that subject age and core knowledge about predators additionally moderate the acquisition of novel call-referent associations.

Preliminary evidence for one-trial social learning of vervet monkey alarm calling

How do non-human primates learn to use their alarm calls? Social learning is a promising candidate, but its role in the acquisition of meaning and call usage has not been studied systematically, neither during ontogeny nor in adulthood. To investigate the role of social learning in alarm call comprehension and use, we exposed groups of wild vervet monkeys to two unfamiliar animal models in the presence or absence of conspecific alarm calls. To assess the learning outcome of these experiences, we then presented the models for a second time to the same monkeys, but now without additional alarm call information. In subjects previously exposed in conjunction with alarm calls, we found heightened predator inspection compared to control subjects exposed without alarm calls, indicating one-trial social learning of 'meaning'. Moreover, some juveniles (but not adults) produced the same alarm calls they heard during the initial exposure whereas the authenticity of the models had an additional effect. Our experiment provides preliminary evidence that, in non-human primates, call meaning can be acquired by one-trail social learning but that subject age and core knowledge about predators additionally moderate the acquisition of novel call-referent associations.

Genetic Parameters of Diameter Growth Dynamics in Norway Spruce Clones

The breeding of Norway spruce in northern Europe has substantially contributed to the production of high-quality wood. The vegetative propagation of robust elite clones could help to sustain the provision of high-quality timber in the face of changing climates. For the adequate evaluation of genetic gains, the altered tree growth dynamics of the clones need to be understood, yet essential information about the long-term growth dynamics of improvedboreal trees is still lacking. We examined a 50-year-old clonal plantation in Latvia to distinguish the clonal effects on diameter growth function parameters and estimate the genetic parameters. A mixed-effect modelling approach was used, in which the clones were applied as random effects on the parameters of the Chapman–Richard equation. All model parameters showed significant variance in the genotypic coefficients of variation CVg which ranged between 11.0 and 17.1%, with the highest being for the growth rate. The heritability (H2) of the diameter at breast height (DBH) reached 0.35 at the age of 40, while CVg decreased from 12.9% to 7.8% between the ages of 20 and 45. Age–age genotypic correlations were positive and were strong or very strong (>0.76). The realised genetic gain varied from −6.3 to +24.0% around the trial mean. A substantial improvement in DBH was indicated when elite clones were selected for vegetative propagation based not only on early measurements, but also considering the genetic variance in the model parameters.

Increased mortality of tropical tree seedlings during the extreme 2015-16 El Niño

As extreme climate events are predicted to become more frequent because of global climate change, understanding their impacts on natural systems is crucial. Tropical forests are vulnerable to droughts associated with extreme El Niño events. However, little is known about how tropical seedling communities respond to El Niño-related droughts, even though patterns of seedling survival shape future forest structure and diversity. Using long-term data from eight tropical moist forests spanning a rainfall gradient in central Panama, we show that community-wide seedling mortality increased by 11% during the extreme 2015-16 El Niño, with mortality increasing most in drought-sensitive species and in wetter forests. These results indicate that severe El Niño-related droughts influence understory dynamics in tropical forests, with effects varying both within and across sites. Our findings suggest that predicted increases in the frequency of extreme El Niño events will alter tropical plant communities through their effects on early life stages.

Species and Phenotypic Distribution Models Reveal Population Differentiation in Ethiopian Indigenous Chickens

Smallholder poultry production dominated by indigenous chickens is an important source of livelihoods for most rural households in Ethiopia. The long history of domestication and the presence of diverse agroecologies in Ethiopia create unique opportunities to study the effect of environmental selective pressures. Species distribution models (SDMs) and Phenotypic distribution models (PDMs) can be applied to investigate the relationship between environmental variation and phenotypic differentiation in wild animals and domestic populations. In the present study we used SDMs and PDMs to detect environmental variables related with habitat suitability and phenotypic differentiation among nondescript Ethiopian indigenous chicken populations. 34 environmental variables (climatic, soil, and vegetation) and 19 quantitative traits were analyzed for 513 adult chickens from 26 populations. To have high variation in the dataset for phenotypic and ecological parameters, animals were sampled from four spatial gradients (each represented by six to seven populations), located in different climatic zones and geographies. Three different ecotypes are proposed based on correlation test between habitat suitability maps and phenotypic clustering of sample populations. These specific ecotypes show phenotypic differentiation, likely in response to environmental selective pressures. Nine environmental variables with the highest contribution to habitat suitability are identified. The relationship between quantitative traits and a few of the environmental variables associated with habitat suitability is non-linear. Our results highlight the benefits of integrating species and phenotypic distribution modeling approaches in characterization of livestock populations, delineation of suitable habitats for specific breeds, and understanding of the relationship between ecological variables and quantitative traits, and underlying evolutionary processes.

Common Buzzards wintering strategies as an effect of weather conditions and geographic barriers

AIM

Migration is a constantly changing adaptation due to the climate condition evolution. The struggle for surviving during harsh winter season is different across Europe, being more complex toward the inner parts of the continent. The current approach explores the Common Buzzard number variation during the cold season and the climatic predictors of birds of prey wintering movements in relation to the possible influences of the Carpathian Mountains, which may act as a geographical barrier providing shelter from cold air outbreak from north and northeast of the continent.

LOCATION

Romania (45°N25°E).

TAXON

Birds of Prey.

METHODS

We applied a GLMM to investigate the relation between continental and local climatic factors with the number of Common Buzzard observations in two regions. The first region is located inside the Carpathian Arch and the other one outside, east of this large mountains chain.

RESULTS

The Common Buzzard numbers wintering Eastern from the Carpathian Mountains are highly influenced by AO (Z = 2.87, p < .05%), while those wintering western are influenced by NAO (Z = 2.17, p < .05%). This is the first proof of separating influences for biodiversity of AO and NAO at continental scale, outlining the influence limit placed over the Eastern Carpathian Mountains.

MAIN CONCLUSIONS

The Carpathian Mountains act like a geographic barrier, separating the wintering Common Buzzard populations from both sides of the mountain range. While the high number of individuals in Moldova is related to their eastern and northeastern Europe origins, in Transylvania the large number of individuals observed is related to the more sheltered characteristics of the region attracting individuals from central Europe. Also, since Transylvania region is well sheltered during cold air outbreak, it represents a more favorable region for wintering. From this point of view, we can consider that the Carpathian Mountains are a geographic barrier for wintering birds of prey.

Warming-driven shifts in ecological control of fish communities in a large northern Chinese lake over 66 years

Warming, land-use change, and habitat loss are three major threats to aquatic biodiversity worldwide under the influences of anthropogenic disturbances. Positive feedback between warming and bottom-up regulation may cause irreversible ecological regime shifts. Threshold dynamics of interspecific interactions have been rarely studied in freshwater fish communities using threshold community models. Here we use 66 years (1950-2015) of data to link four ecological regime shifts of 9-species fish communities to climatic and land use changes in Lake Hulun, the largest freshwater lake of Northern China. Overfishing caused the collapse of piscivorous fish populations and an ecological regime shift of Lake Hulun in the late 1950s. The first recorded algal bloom of Lake Hulun took place in 1986, with accelerated warming and rapid increases in livestock grazing. The dominance of planktivorous minnow populations reduced fish biodiversity in a nonlinear, threshold manner when annual mean ambient temperature was >0.12 °C. Multivariate environmental vector regression demonstrated that warming, eutrophication, and water-storage reduction (i.e., habitat loss) were related to three ecological regime shifts of Lake Hulun from 1960 to 2015. Multivariate autoregressive models (MAR) did not detect predation by piscivorous fish in Lake Hulun after 1960. Threshold MAR models indicated that dominant minnow populations and other prey fish populations switched from top-down to bottom-up control during the 1980s. Sustained positive feedback between warming, the dominance of planktivorous fish populations, and bottom-up regulation caused predator-prey role reversal, and probably resulted in three regime shifts of Lake Hulun over 56 years. This study provides a comprehensive analysis of ecological regime shifts in Hulun Lake fish communities, and has potential implications for fish species living in similar environments that are subject to global warming, land-use changes, and overfishing.

Unexpected Effects of Local Management and Landscape Composition on Predatory Mites and Their Food Resources in Vineyards

Viticultural practices and landscape composition are the main drivers influencing biological pest control in vineyards. Predatory mites, mainly phytoseiid (Phytoseiidae) and tydeoid mites (Tydeidae), are important to control phytophagous mites (Tetranychidae and Eriophyidae) on vines. In the absence of arthropod prey, pollen is an important food source for predatory mites. In 32 paired vineyards located in Burgenland/Austria, we examined the effect of landscape composition, management type (organic/integrated), pesticide use, and cover crop diversity of the inter-row on the densities of phytoseiid, tydeoid, and phytophagous mites. In addition, we sampled pollen on vine leaves. Typhlodromus pyri Scheuten was the main phytoseiid mite species and Tydeus goetzi Schruft the main tydeoid species. Interestingly, the area-related acute pesticide toxicity loading was higher in organic than in integrated vineyards. The densities of phytoseiid and tydeoid mites was higher in integrated vineyards and in vineyards with spontaneous vegetation. Their population also profited from an increased viticultural area at the landscape scale. Eriophyoid mite densities were extremely low across all vineyards and spider mites were absent. Biological pest control of phytophagous mites benefits from less intensive pesticide use and spontaneous vegetation cover in vineyard inter-rows, which should be considered in agri-environmental schemes.

Biological and statistical interpretation of size-at-age, mixed-effects models of growth

The differences in life-history traits and processes between organisms living in the same or different populations contribute to their ecological and evolutionary dynamics. We developed mixed-effect model formulations of the popular size-at-age von Bertalanffy and Gompertz growth functions to estimate individual and group variation in body growth, using as a model system four freshwater fish populations, where tagged individuals were sampled for more than 10 years. We used the software Template Model Builder to estimate the parameters of the mixed-effect growth models. Tests on data that were not used to estimate model parameters showed good predictions of individual growth trajectories using the mixed-effects models and starting from one single observation of body size early in life; the best models had R 2 > 0.80 over more than 500 predictions. Estimates of asymptotic size from the Gompertz and von Bertalanffy models were not significantly correlated, but their predictions of size-at-age of individuals were strongly correlated (r > 0.99), which suggests that choosing between the best models of the two growth functions would have negligible effects on the predictions of size-at-age of individuals. Model results pointed to size ranks that are largely maintained throughout the lifetime of individuals in all populations.

Performance, motivation, and enjoyment in young female basketball players: An interdisciplinary approach

The purpose of this study was to examine the variation in accumulated basketball training experience, body size, functional performance, deliberate practice motivation, achievement and competitiveness motivation and sources of enjoyment among young female basketball players, partitioning the potential variation by individuals´ biological characteristics (menarche status) and contextual characteristics (competitive age group and competitive level). We considered 114 adolescent female basketball players aged 14.3 (1.8) years. We used multilevel regression and poststratification estimations. The adolescent female basketball players selected for state-level had more accumulated experience, were taller and with better functional performance. Conditional on the data, youth female coaches tend to value (probably overvalue) size and function when selecting/promoting players, even at early age groups, likely contributing to an overrepresentation of early maturing girls in at early age groups. Players from club- and state-level were similarly highly motivated for deliberate practice and to achievement. Only for competitiveness, state-level players had higher values than club level players. The sources of enjoyment were influenced by context (competitive levels) for self-referenced competencies and others-referenced competencies. Structured programs of training and competition in youth female basketball provide a nurturing environment for the development of players´ engagement and commitment to training and excellence attainment.

A Meta-analysis of Plant Interaction Networks Reveals Competitive Hierarchies as well as Facilitation and Intransitivity

The extent to which competitive interactions and niche differentiation structure communities has been highly controversial. To quantify evidence for key features of plant community structure, I recharacterized published data from interaction experiments as networks of competitive and facilitative interactions. I measured the network structure of 31 woody and herbaceous communities, including the intensity, distribution, and diversity of interactions at the species-pair and community levels to determine the generality of competition, winner-loser relationships, and unequal interaction allocation. I developed novel methodology using meta-analysis to incorporate interaction uncertainty into estimates of structural metrics among independent networks. Plant communities were competitive, but intraspecific interactions were sometimes more intense than interspecific interactions. On the whole, interactions were imbalanced and communities were transitive. However, facilitation, balanced interactions, and intransitivity were common in individual communities. Synthesizing network metrics using meta-analysis is an original approach with which to generalize community structure in a systematic way.

Detecting Consensus Emergence in Organizational Multilevel Data: Power Simulations

Theories suggest that groups within organizations often develop shared values, beliefs, affect, behaviors, or agreed-on routines; however, researchers rarely study predictors of consensus emergence over time. Recently, a multilevel-methods approach for detecting and studying emergence in organizational field data has been described. This approach—the consensus emergence model—builds on an extended three-level multilevel model. Researchers planning future studies based on the consensus emergence model need to consider (a) sample size characteristics required to detect emergence effects with satisfactory statistical power and (b) how the distribution of the overall sample size across the levels of the multilevel model influences power. We systematically address both issues by conducting a power simulation for detecting main and moderating effects involving consensus emergence under a variety of typical research scenarios and provide an R-based tool that readers can use to estimate power. Our discussion focuses on the future use and development of multilevel methods for studying emergence in organizational research.

A nonlinear mixed-effects modeling approach for ecological data: Using temporal dynamics of vegetation moisture as an example

Increasingly, often ecologist collects data with nonlinear trends, heterogeneous variances, temporal correlation, and hierarchical structure. Nonlinear mixed-effects models offer a flexible approach to such data, but the estimation and interpretation of these models present challenges, partly associated with the lack of worked examples in the ecological literature.We illustrate the nonlinear mixed-effects modeling approach using temporal dynamics of vegetation moisture with field data from northwestern Patagonia. This is a Mediterranean-type climate region where modeling temporal changes in live fuel moisture content are conceptually relevant (ecological theory) and have practical implications (fire management). We used this approach to answer whether moisture dynamics varies among functional groups and aridity conditions, and compared it with other simpler statistical models. The modeling process is set out "step-by-step": We start translating the ideas about the system dynamics to a statistical model, which is made increasingly complex in order to include different sources of variability and correlation structures. We provide guidelines and R scripts (including a new self-starting function) that make data analyses reproducible. We also explain how to extract the parameter estimates from the R output.Our modeling approach suggests moisture dynamic to vary between grasses and shrubs, and between grasses facing different aridity conditions. Compared to more classical models, the nonlinear mixed-effects model showed greater goodness of fit and met statistical assumptions. While the mixed-effects approach accounts for spatial nesting, temporal dependence, and variance heterogeneity; the nonlinear function allowed to model the seasonal pattern.Parameters of the nonlinear mixed-effects model reflected relevant ecological processes. From an applied perspective, the model could forecast the time when fuel moisture becomes critical to fire occurrence. Due to the lack of worked examples for nonlinear mixed-effects models in the literature, our modeling approach could be useful to diverse ecologists dealing with complex data.

Predator-prey dynamics of bald eagles and glaucous-winged gulls at Protection Island, Washington, USA

Bald eagle (Haliaeetus leucocephalus) populations in North America rebounded in the latter part of the twentieth century, the result of tightened protection and outlawing of pesticides such as DDT. An unintended consequence of recovery may be a negative impact on seabirds. During the 1980s, few bald eagles disturbed a large glaucous-winged gull (Larus glaucescens) colony on Protection Island, Washington, USA, in the Salish Sea. Breeding gull numbers in this colony rose nearly 50% during the 1980s and early 1990s. Beginning in the 1990s, a dramatic increase in bald eagle activity ensued within the colony, after which began a significant decline in gull numbers.To examine whether trends in the gull colony could be explained by eagle activity, we fit a Lotka-Volterra-type predator-prey model to gull nest count data and Washington State eagle territory data collected in most years between 1980 and 2016. Both species were assumed to grow logistically in the absence of the other.The model fits the data with generalized R 2 = 0.82, supporting the hypothesis that gull dynamics were due largely to eagle population dynamics.Point estimates of the model parameters indicated approach to stable coexistence. Within the 95% confidence intervals for the parameters, however, 11.0% of bootstrapped parameter vectors predicted gull colony extinction.Our results suggest that the effects of bald eagle activity on the dynamics of a large gull colony were explained by a predator-prey relationship that included the possibility of coexistence but also the possibility of gull colony extinction. This study serves as a cautionary exploration of the future, not only for gulls on Protection Island, but for other seabirds in the Salish Sea. Managers should monitor numbers of nests in seabird colonies as well as eagle activity within colonies to document trends that may lead to colony extinction.

EStimating Contaminants tRansfers Over Complex food webs (ESCROC): An innovative Bayesian method for estimating POP's biomagnification in aquatic food webs

Pollution greatly impacts ecosystems health and associated ecological functions. Persistent Organic Pollutants (POPs) are among the most studied contaminants due to their persistence, bioaccumulation, and toxicity potential. Biomagnification is often described using the estimation of a Trophic Magnification Factor (TMF). This estimate is based on the relationship between contamination levels of the species and their trophic level. However, while the estimation can be significantly biased in relation to multiple sources of uncertainty (e.g. species physiology, measurement errors, food web complexity), usual TMF estimation methods typically do not allow accounting for these potential biases. More accurate and reliable assessment tool of TMFs and their associated uncertainty are therefore needed in order to appropriately guide chemical pollution management. The present work proposes a relevant and innovative TMF estimation method accounting for its many variability sources. The ESCROC model (EStimating Contaminants tRansfers Over Complex food webs), which is implemented in a Bayesian framework, allows for a more reliable and rigorous assessment of contaminants trophic magnification, in addition to accurate estimations of isotopes trophic enrichment factors and their associated uncertainties in food webs. Similar to classical mixing models used in food web investigations, ECSROC computes diet composition matrices using isotopic composition data while accounting for contamination data, leading to more robust food web descriptions. As a demonstration of the practical application of the model, ESCROC was implemented to revisit the trophic biomagnification of 5 polyfluoroalkyl substances (PFAS) in a complex estuarine food web (the Gironde, SW France). In addition to the TMF estimate and 95% confidence intervals, the model provided biomagnification probabilities associated to the investigated contaminants-for instance, 92% in the case of perfluorooctane sulfonate (PFOS)-that can be interpreted in terms of risk assessment in a precautionary approach, which should prove useful to environmental managers.

Parameter estimation using randomized phases in an integrated assessment model for Antarctic krill

An integrated model assessing the status and productivity of Antarctic krill (Euphausia superba, hereafter krill) was configured to estimate different subsets of 118 potentially estimable parameters in alternative configurations. We fixed the parameters that were not estimated in any given configuration at pre-specified values. The model was fitted to over forty years of fisheries and survey data for krill in Subarea 48.1, a statistical reporting area around the Antarctic Peninsula used by the Commission for the Conservation of Antarctic Marine Living Resources (CCAMLR). The number of estimated parameters was gradually increased across model configurations. Configurations that estimated more parameters fitted the data better, but the order in which the parameters were estimated became more important in finding the best fit. Twenty-two configurations estimating from 48 to 107 parameters were able to obtain an invertible Hessian matrix that was subsequently used to estimate parameter uncertainty. Parameter uncertainties calculated using asymptotic approximation around the maximum likelihood estimates were often larger than uncertainties based on Markov chain Monte Carlo sampling for the same parameters. Diagnostics applied to MCMC samples in the best model of each configuration that obtained an invertible Hessian indicated that the most highly parameterized configurations did not reach stationary distributions. A 96-parameter configuration was the best fitting model of those that passed the MCMC diagnostics. The ΔAIC and ΔBIC scores indicated essentially no support relative to the best model for the alternative models that also passed MCMC diagnostics. Simulated data using the configurations as operating models showed that while all configurations passed "self-tests" for spawning biomass and recruitment, there was a small negative bias due to model penalties in the fishing mortality estimates for years with the highest fishing mortalities. "Cross-tests" of configurations that estimated different parameters often differed from the operating model values.

Rare genotype advantage promotes survival and genetic diversity of a tropical palm

Negative density dependence, where survival decreases as density increases, is a well-established driver of species diversity at the community level, but the degree to which a similar process might act on the density or frequency of genotypes within a single plant species to maintain genetic diversity has not been well studied in natural systems. In this study, we determined the maternal genotype of naturally dispersed seeds of the palm Oenocarpus bataua within a tropical forest in northwest Ecuador, tracked the recruitment of each seed, and assessed the role of individual-level genotypic rarity on survival. We demonstrate that negative frequency-dependent selection within this species conferred a survival advantage to rare maternal genotypes and promoted population-level genetic diversity. The strength of the observed rare genotype survival advantage was comparable to the effect of conspecific density regardless of genotype. These findings corroborate an earlier, experimental study and implicate negative frequency-dependent selection of genotypes as an important, but currently underappreciated, determinant of plant recruitment and within-species genetic diversity. Incorporating intraspecific genetic variation into studies and theory of forest dynamics may improve our ability to understand and manage forests, and the processes that maintain their diversity.

No-U-turn sampling for fast Bayesian inference in ADMB and TMB: Introducing the adnuts and tmbstan R packages

Statistical inference is a widely-used, powerful tool for learning about natural processes in diverse fields. The statistical software platforms AD Model Builder (ADMB) and Template Model Builder (TMB) are particularly popular in the ecological literature, where they are typically used to perform frequentist inference of complex models. However, both lack capabilities for flexible and efficient Markov chain Monte Carlo (MCMC) integration. Recently, the no-U-turn sampler (NUTS) MCMC algorithm has gained popularity for Bayesian inference through the software Stan because it is efficient for high dimensional, complex hierarchical models. Here, we introduce the R packages adnuts and tmbstan, which provide NUTS sampling in parallel and interactive diagnostics with ShinyStan. The ADMB source code was modified to provide NUTS, while TMB models are linked directly into Stan. We describe the packages, provide case studies demonstrating their use, and contrast performance against Stan. For TMB models, we show how to test the accuracy of the Laplace approximation using NUTS. For complex models, the performance of ADMB and TMB was typically within +/- 50% the speed of Stan. In one TMB case study we found inaccuracies in the Laplace approximation, potentially leading to biased inference. adnuts provides a new method for estimating hierarchical ADMB models which previously were infeasible. TMB users can fit the same model in both frequentist and Bayesian paradigms, including using NUTS to test the validity of the Laplace approximation of the marginal likelihood for arbitrary subsets of parameters. These software developments extend the available statistical methods of the ADMB and TMB user base with no additional effort by the user.

Models to describe the thermal development rates of Cycloneda sanguinea L. (Coleoptera: Coccinelidae)

Precise estimates of the lower (T_min) and higher (T_max) thermal thresholds as well as the temperature range that provides optimum performance (T_opt) enable to obtain the desired number of individuals in conservation systems, rearing and release of natural enemies. In this study, the relationship between the development rates of Cycloneda sanguinea L. (Coleoptera: Coccinelidae) and temperature was described using non-linear models developed by Analytis, Brière, Lactin, Lamb, Logan and Sharpe & DeMichele. There were differences between the models, considering the estimates of the parameters T_min, T_max , and T_opt. All of the tested models were able to describe non-linear responses involving the development rates of C. sanguinea at constant temperatures. Lactin and Lamb gave the highest z weight for egg, while Analytis, Sharpe & DeMichele and Brière gave the highest values for larvae and pupae. The more realistic T_opt estimated by the models varied from 29° to 31°C for egg, 27-28 °C for larvae and 28-29 °C for pupae. The Logan, Lactin and Analytis models estimated the T_max for egg, larvae and pupae to be approximately 34 °C, while the T_min estimated by the Analytis model was 16 °C for larvae and pupae. The information generated by our research will contribute towards improving the rearing and release of C. sanguinea in biological control programs, accurately controlling the rate of development in laboratory conditions or even scheduling the most favourable this species' release.

A framework for estimating the determinants of spatial and temporal variation in vital rates and inferring the occurrence of unobserved extreme events

We develop a general framework that combines long-term tag-recapture data and powerful statistical and modelling techniques to investigate how population, environmental and climate factors determine variation in vital rates and population dynamics in an animal species, using as a case study the population of brown trout living in Upper Volaja (Western Slovenia). This population has been monitored since 2004. Upper Volaja is a sink, receiving individuals from a source population living above a waterfall. We estimate the numerical contribution of the source population on the sink population and test the effects of temperature, population density and extreme events on variation in vital rates among 2647 individually tagged brown trout. We found that individuals dispersing downstream from the source population help maintain high population densities in the sink population despite poor recruitment. The best model of survival for individuals older than juveniles includes additive effects of birth cohort and sampling occasion. Fast growth of older cohorts and higher population densities in 2004-2005 suggest very low population densities in the late 1990s, which we hypothesize were caused by a flash flood that strongly reduced population size and created the habitat conditions for faster individual growth and transient higher population densities after the extreme event.

Multinomial analysis of behavior: statistical methods

Behavioral ecologists frequently use observational methods, such as instantaneous scan sampling, to record the behavior of animals at discrete moments in time. We develop and apply multilevel, multinomial logistic regression models for analyzing such data. These statistical methods correspond to the multinomial character of the response variable while also accounting for the repeated observations of individuals that characterize behavioral datasets. Correlated random effects potentially reveal individual-level trade-offs across behaviors, allowing for models that reveal the extent to which individuals who regularly engage in one behavior also exhibit relatively more or less of another behavior. Using an example dataset, we demonstrate the estimation of these models using Hamiltonian Monte Carlo algorithms, as implemented in the RStan package in the R statistical environment. The supplemental files include a coding script and data that demonstrate auxiliary functions to prepare the data, estimate the models, summarize the posterior samples, and generate figures that display model predictions. We discuss possible extensions to our approach, including models with random slopes to allow individual-level behavioral strategies to vary over time and the need for models that account for temporal autocorrelation. These models can potentially be applied to a broad class of statistical analyses by behavioral ecologists, focusing on other polytomous response variables, such as behavior, habitat choice, or emotional states.

Alien plant invasions of protected areas in Java, Indonesia

Alien plants are invading protected areas worldwide, but there is little information from tropical Asia. Java has the longest record of human occupation in Asia and today supports 145 m people. Remnants of natural ecosystems survive in 12 small National Parks surrounded by dense human populations, making them highly vulnerable to invasions. We surveyed eight of these, along a rainfall gradient from lowland rainforest with >3000 mm annual rainfall to savanna with <1500 mm, and a 0-3158 m altitudinal gradient, using 403 10 × 5 m plots along trails. We found 67 invasive alien plant species, of which 33 occurred in only one park and two (Chromolaena odorata and Lantana camara) in all. Historical factors relating to plant introduction appeared to be as important as environmental factors in determining which species occurred in which park, while within parks canopy cover and altitude were generally most influential. Spread away from trails was only evident in open habitats, including natural savannas in Baluran National Park, threatened by invasion of Acacia nilotica. Existing control attempts for invasive aliens are reactive, localized, and intermittent, and insufficient resources are currently available for the early detection, prompt action, and continued monitoring that are needed.

Reduced abundance and earlier collection of bumble bee workers under intensive cultivation of a mass-flowering prairie crop

One of the most commonly seeded crops in Canada is canola, a cultivar of oilseed rape (Brassica napus). As a mass‐flowering crop grown intensively throughout the Canadian Prairies, canola has the potential to influence pollinator success across tens of thousands of square kilometers of cropland. Bumble bees (Bombus sp.) are efficient pollinators of many types of native and crop plants. We measured the influence of this mass‐flowering crop on the abundance and phenology of bumble bees, and on another species of social bee (a sweat bee; Halictus rubicundus), by continuously deploying traps at different levels of canola cultivation intensity, spanning the start and end of canola bloom. Queen bumble bees were more abundant in areas with more canola cover, indicating that this crop is attractive to queens. However, bumble bee workers were significantly fewer in these locations later in the season, suggesting reduced colony success. The median collection dates of workers of three bumble bee species were earlier near canola fields, suggesting a dynamic response of colonies to the increased floral resources. Different species experienced this shift to different extents. The sweat bee was not affected by canola cultivation intensity. Our findings suggest that mass‐flowering crops such as canola are attractive to bumble bee queens and therefore may lead to higher rates of colony establishment, but also that colonies established near this crop may be less successful. We propose that the effect on bumble bees can be mitigated by spacing the crop more evenly with respect to alternate floral resources.

Trade-offs between accuracy and interpretability in von Bertalanffy random-effects models of growth

Better understanding of variation in growth will always be an important problem in ecology. Individual variation in growth can arise from a variety of processes; for example, individuals within a population vary in their intrinsic metabolic rates and behavioral traits, which may influence their foraging dynamics and access to resources. However, when adopting a growth model, we face trade-offs between model complexity, biological interpretability of parameters, and goodness of fit. We explore how different formulations of the von Bertalanffy growth function (vBGF) with individual random effects and environmental predictors affect these trade-offs. In the vBGF, the growth of an organism results from a dynamic balance between anabolic and catabolic processes. We start from a formulation of the vBGF that models the anabolic coefficient (q) as a function of the catabolic coefficient (k), a coefficient related to the properties of the environment (γ) and a parameter that determines the relative importance of behavior and environment in determining growth (ψ). We treat the vBGF parameters as a function of individual random effects and environmental variables. We use simulations to show how different functional forms and individual or group variability in the growth function's parameters provide a very flexible description of growth trajectories. We then consider a case study of two fish populations of Salmo marmoratus and Salmo trutta to test the goodness of fit and predictive power of the models, along with the biological interpretability of vBGF's parameters when using different model formulations. The best models, according to AIC, included individual variability in both k and γ and cohort as predictor of growth trajectories, and are consistent with the hypothesis that habitat selection is more important than behavioral and metabolic traits in determining lifetime growth trajectories of the two fish species. Model predictions of individual growth trajectories were largely more accurate than predictions based on mean size-at-age of fish. Our method shares information across individuals, and thus, for both fish populations investigated, allows using a single measurement early in the life of individual fish or cohort to obtain accurate predictions of lifetime individual or cohort size-at-age.

Negative effects of pesticides on wild bee communities can be buffered by landscape context

Wild bee communities provide underappreciated but critical agricultural pollination services. Given predicted global shortages in pollination services, managing agroecosystems to support thriving wild bee communities is, therefore, central to ensuring sustainable food production. Benefits of natural (including semi-natural) habitat for wild bee abundance and diversity on farms are well documented. By contrast, few studies have examined toxicity of pesticides on wild bees, let alone effects of farm-level pesticide exposure on entire bee communities. Whether beneficial natural areas could mediate effects of harmful pesticides on wild bees is also unknown. Here, we assess the effect of conventional pesticide use on the wild bee community visiting apple (Malus domestica) within a gradient of percentage natural area in the landscape. Wild bee community abundance and species richness decreased linearly with increasing pesticide use in orchards one year after application; however, pesticide effects on wild bees were buffered by increasing proportion of natural habitat in the surrounding landscape. A significant contribution of fungicides to observed pesticide effects suggests deleterious properties of a class of pesticides that was, until recently, considered benign to bees. Our results demonstrate extended benefits of natural areas for wild pollinators and highlight the importance of considering the landscape context when weighing up the costs of pest management on crop pollination services.

Fast fitting of non-Gaussian state-space models to animal movement data via Template Model Builder

State-space models (SSM) are often used for analyzing complex ecological processes that are not observed directly, such as marine animal movement. When outliers are present in the measurements, special care is needed in the analysis to obtain reliable location and process estimates. Here we recommend using the Laplace approximation combined with automatic differentiation (as implemented in the novel R package Template Model Builder; TMB) for the fast fitting of continuous-time multivariate non-Gaussian SSMs. Through Argos satellite tracking data, we demonstrate that the use of continuous-time t-distributed measurement errors for error-prone data is more robust to outliers and improves the location estimation compared to using discretized-time t-distributed errors (implemented with a Gibbs sampler) or using continuous-time Gaussian errors (as with the Kalman filter). Using TMB, we are able to estimate additional parameters compared to previous methods, all without requiring a substantial increase in computational time. The model implementation is made available through the R package argosTrack.

Niche Partitioning of Feather Mites within a Seabird Host, Calonectris borealis

According to classic niche theory, species can coexist in heterogeneous environments by reducing interspecific competition via niche partitioning, e.g. trophic or spatial partitioning. However, support for the role of competition on niche partitioning remains controversial. Here, we tested for spatial and trophic partitioning in feather mites, a diverse and abundant group of arthropods. We focused on the two dominant mite species, Microspalax brevipes and Zachvatkinia ovata, inhabiting flight feathers of the Cory’s shearwater, Calonectris borealis. We performed mite counts across and within primary and tail feathers on free-living shearwaters breeding on an oceanic island (Gran Canaria, Canary Islands). We then investigated trophic relationships between the two mite species and the host using stable isotope analyses of carbon and nitrogen on mite tissues and potential host food sources. The distribution of the two mite species showed clear spatial segregation among feathers; M. brevipes showed high preference for the central wing primary feathers, whereas Z. ovata was restricted to the two outermost primaries. Morphological differences between M. brevipes and Z. ovata support an adaptive basis for the spatial segregation of the two mite species. However, the two mites overlap in some central primaries and statistical modeling showed that Z. ovata tends to outcompete M. brevipes. Isotopic analyses indicated similar isotopic values for the two mite species and a strong correlation in carbon signatures between mites inhabiting the same individual host suggesting that diet is mainly based on shared host-associated resources. Among the four candidate tissues examined (blood, feather remains, skin remains and preen gland oil), we conclude that the diet is most likely dominated by preen gland oil, while the contribution of exogenous material to mite diets is less marked. Our results indicate that ongoing competition for space and resources plays a central role in structuring feather mite communities. They also illustrate that symbiotic infracommunities are excellent model systems to study trophic ecology, and can improve our understanding of mechanisms of niche differentiation and species coexistence.

Partitioning detectability components in populations subject to within-season temporary emigration using binomial mixture models

Detectability of individual animals is highly variable and nearly always < 1; imperfect detection must be accounted for to reliably estimate population sizes and trends. Hierarchical models can simultaneously estimate abundance and effective detection probability, but there are several different mechanisms that cause variation in detectability. Neglecting temporary emigration can lead to biased population estimates because availability and conditional detection probability are confounded. In this study, we extend previous hierarchical binomial mixture models to account for multiple sources of variation in detectability. The state process of the hierarchical model describes ecological mechanisms that generate spatial and temporal patterns in abundance, while the observation model accounts for the imperfect nature of counting individuals due to temporary emigration and false absences. We illustrate our model's potential advantages, including the allowance of temporary emigration between sampling periods, with a case study of southern red-backed salamanders Plethodon serratus. We fit our model and a standard binomial mixture model to counts of terrestrial salamanders surveyed at 40 sites during 3-5 surveys each spring and fall 2010-2012. Our models generated similar parameter estimates to standard binomial mixture models. Aspect was the best predictor of salamander abundance in our case study; abundance increased as aspect became more northeasterly. Increased time-since-rainfall strongly decreased salamander surface activity (i.e. availability for sampling), while higher amounts of woody cover objects and rocks increased conditional detection probability (i.e. probability of capture, given an animal is exposed to sampling). By explicitly accounting for both components of detectability, we increased congruence between our statistical modeling and our ecological understanding of the system. We stress the importance of choosing survey locations and protocols that maximize species availability and conditional detection probability to increase population parameter estimate reliability.

Hidden semi-Markov models reveal multiphasic movement of the endangered Florida panther

Animals must move to find food and mates, and to avoid predators; movement thus influences survival and reproduction, and ultimately determines fitness. Precise description of movement and understanding of spatial and temporal patterns as well as relationships with intrinsic and extrinsic factors is important both for theoretical and applied reasons. We applied hidden semi-Markov models (HSMM) to hourly geographic positioning system (GPS) location data to understand movement patterns of the endangered Florida panther (Puma concolor coryi) and to discern factors influencing these patterns. Three distinct movement modes were identified: (1) Resting mode, characterized by short step lengths and turning angles around 180(o); (2) Moderately active (or intermediate) mode characterized by intermediate step lengths and variable turning angles, and (3) Traveling mode, characterized by long step lengths and turning angles around 0(o). Males and females, and females with and without kittens, exhibited distinctly different movement patterns. Using the Viterbi algorithm, we show that differences in movement patterns of male and female Florida panthers were a consequence of sex-specific differences in diurnal patterns of state occupancy and sex-specific differences in state-specific movement parameters, whereas the differences between females with and without dependent kittens were caused solely by variation in state occupancy. Our study demonstrates the use of HSMM methodology to precisely describe movement and to dissect differences in movement patterns according to sex, and reproductive status.

Determining individual variation in growth and its implication for life-history and population processes using the empirical Bayes method

The differences in demographic and life-history processes between organisms living in the same population have important consequences for ecological and evolutionary dynamics. Modern statistical and computational methods allow the investigation of individual and shared (among homogeneous groups) determinants of the observed variation in growth. We use an Empirical Bayes approach to estimate individual and shared variation in somatic growth using a von Bertalanffy growth model with random effects. To illustrate the power and generality of the method, we consider two populations of marble trout Salmo marmoratus living in Slovenian streams, where individually tagged fish have been sampled for more than 15 years. We use year-of-birth cohort, population density during the first year of life, and individual random effects as potential predictors of the von Bertalanffy growth function's parameters k (rate of growth) and (asymptotic size). Our results showed that size ranks were largely maintained throughout marble trout lifetime in both populations. According to the Akaike Information Criterion (AIC), the best models showed different growth patterns for year-of-birth cohorts as well as the existence of substantial individual variation in growth trajectories after accounting for the cohort effect. For both populations, models including density during the first year of life showed that growth tended to decrease with increasing population density early in life. Model validation showed that predictions of individual growth trajectories using the random-effects model were more accurate than predictions based on mean size-at-age of fish.

Somatic growth is a crucial determinant of ecological and evolutionary dynamics, since larger organisms often have higher survival and reproductive success. Size may be the result of intrinsic (i.e. genetic), environmental (temperature, food), and social (competition with conspecifics) factors and interaction between them. Knowing the contribution of intrinsic, environmental, and social factors will improve our understanding of individual population dynamics, help conservation and management of endangered species, and increase our ability to predict future growth trajectories of individuals and populations. The latter goal is also relevant for humans, since predicting future growth of newborns may help identify early pathologies that occur later in life. However, teasing apart the contribution of individual and environmental factors requires powerful and efficient statistical methods, as well as biological insights and the use of longitudinal data. We developed a novel statistical approach to estimate and separate the contribution of intrinsic and environmental factors to lifetime growth trajectories, and generate hypotheses concerning the life-history strategies of organisms. Using two fish populations as a case study, we show that our method predicts future growth of organisms with substantially greater accuracy than using historical information on growth at the population level, and help us identify year-class effects, probably associated with climatic vagaries, as the most important environmental determinant of growth.