Overcompensation and phase effects in a cyclic common vole population: between first and second-order cycles

Climate-driven context-dependent structure of population cycles

Multiannual population cycles of small mammals are of interest within population biology. We propose an approach for multidimensional autoregressive (AR) time series and analyse monitoring data on grey-sided voles (Myodes rufocanus) in Japan to investigate one or possibly multiple multiannual cycles that drive population dynamics. Temperature, through modifying rodent communities, is found to be a key factor shaping population dynamics. Warmer areas are the main habitat for other rodent species resulting in low vole abundance/dominance, as opposed to higher vole dominance in colder areas-a pattern associated with the AR structure and population cycle. Vole populations in simple rodent communities exhibit an AR(2) cycle of 2-3 years. In areas with complex rodent communities, vole dynamics follows an AR(4) process and a combination of two cycles with different lengths. The AR structure varies in relatively small spatial scales, thus widening the scope of AR analyses needed. Historically, vole abundance increased in the late 1970s and decreased from the 1980s, with warm winters shown to be associated with the decline of vole abundance in the AR(4) populations. This significant association between the AR order, population dynamics, temperature and rodent community provides insights into the declining trends observed in rodent populations of the Northern Hemisphere.

Are population cycles recovering?

The dynamics of populations of small mammals of Central Siberia was analyzed. The studies were carried out at the Yenisei ecological station "Mirnoye" of the A.N. Severtsov Institute of Ecology and Evolution, Russian Academy of Sciences. The time series analysis was performed by the wavelet transform using the statistical data processing language R. In the 20th century, the dynamics of the population of the community and some of its constituent species (Sorex araneus; S. caecutiens; S. isodon; S. tundrensis; S. minutus; Craseomys rufocanus; Clethrionomys rutilus; Microtus oeconomus; M. agrestis) were characterized by a 4-year periodicity. The type of dynamics changed to noncyclic by the nineties, but by 2022, four species (S. araneus, S. isodon, C. rutilus, and M. oeconomus) and the community as a whole showed a tendency toward recovery of population cycles. The remaining species were characterized by consistently low numbers with irregular low amplitude fluctuations.

Climate variability and density-dependent population dynamics: Lessons from a simple High Arctic ecosystem

Whether the renowned population cycles of small mammals in northern food webs are driven by bottom-up (plant–herbivore) or top-down (predator–prey) interactions is still a debated question but crucial to our understanding of their ecological functions and response to climate change. A long-term study of a graminivorous vole population in an exceptionally simple High Arctic food web allowed us to identify which population dynamics features are present without top-down regulation. Unique features were high-amplitude, noncyclic population fluctuations driven by a combination of stochastic weather events and season-specific density dependence likely arising from plant–herbivore interactions. That such features are not present in more complex food webs points to the importance of top-down regulation in small mammal populations.

Ecologists are still puzzled by the diverse population dynamics of herbivorous small mammals that range from high-amplitude, multiannual cycles to stable dynamics. Theory predicts that this diversity results from combinations of climatic seasonality, weather stochasticity, and density-dependent food web interactions. The almost ubiquitous 3- to 5-y cycles in boreal and arctic climates may theoretically result from bottom-up (plant–herbivore) and top-down (predator–prey) interactions. Assessing, empirically, the roles of such interactions and how they are influenced by environmental stochasticity has been hampered by food web complexity. Here, we take advantage of a uniquely simple High Arctic food web, which allowed us to analyze the dynamics of a graminivorous vole population not subjected to top-down regulation. This population exhibited high-amplitude, noncyclic fluctuations—partly driven by weather stochasticity. However, the predominant driver of the dynamics was overcompensatory density dependence in winter that caused the population to frequently crash. Model simulations showed that the seasonal pattern of density dependence would yield regular 2-y cycles in the absence of stochasticity. While such short cycles have not yet been observed in mammals, they are theoretically plausible if graminivorous vole populations are deterministically bottom-up regulated. When incorporating weather stochasticity in the model simulations, cyclicity became disrupted and the amplitude was increased—akin to the observed dynamics. Our findings contrast with the 3- to 5-y population cycles that are typical of graminivorous small mammals in more complex food webs, suggesting that top-down regulation is normally an important component of such dynamics.

Do human infrastructures shape nest distribution in the landscape depending on individual personality in a farmland bird of prey?

Individuals' distribution across habitats may depend on their personality. Human activities and infrastructures are critical elements of the landscape that may impact the habitat selection process. However, depending on their personality, individuals may respond differently to these unnatural elements. In the present study, we first investigated whether some human infrastructures (buildings, roads and paths) shaped Montagu's harrier nest spatial distribution in the landscape according to female personality (boldness). Second, we tested if the reproductive success of females depended on their boldness and nest location regarding infrastructures. Using a long-term (19 years) dataset, we calculated, for each infrastructure type, the distance from each nest to the nearest infrastructure and the infrastructure density around the nest. We tested the effects of female boldness (bold vs. shy) and its interaction with egg-laying date on these six metrics. Nest location in the landscape depended on female personality and on some human infrastructures: the building density was smaller around nests from shy females than from bold ones. Nest distribution related to other infrastructure metrics did not depend on female boldness. The pattern related to building density is consistent with some habitat choice hypotheses, which are discussed. Path density around nests negatively affected reproductive success regardless of female boldness, and late breeders nested further away from paths than early breeders. Human activities on paths (more common later in the season) could lead to disturbance and a decrease in parental care, reducing reproductive success. Increasing human presence in farmlands implies a need to better understand its impact on population composition, in terms of personality. Our results suggest that individual behavioural differences should be taken into account in studies assessing the effects of human disturbance on animal populations, to propose more appropriate conservation measures.

Quantifying the nature and strength of intraspecific density dependence in Arctic mosquitoes

Processes that change with density are inherent in all populations, yet quantifying density dependence with empirical data remains a challenge. This is especially true for animals recruiting in patchy landscapes because heterogeneity in habitat quality in combination with habitat choice can obscure patterns expected from density dependence. Mosquitoes (Diptera: Culicidae) typically experience strong density dependence when larvae compete for food, however, effects vary across species and contexts. If populations experience intense intraspecific density-dependent mortality then overcompensation can occur, where the number of survivors declines at high densities producing complex endogenous dynamics. To seek generalizations about density dependence in a widespread species of Arctic mosquito, Aedes nigripes, we combined a laboratory experiment, field observations, and modeling approaches. We evaluated alternative formulations of discrete population models and compared best-performing models from our lab study to larval densities from ponds in western Greenland. Survivorship curves from the lab were the best fit by a Hassell model with compensating density dependence (equivalent to a Beverton-Holt model) where peak recruitment ranged from 8 to 80 mosquitoes per liter depending on resource supply. In contrast, our field data did not show a signal of strong density dependence, suggesting that other processes such as predation may lower realized densities in nature, and that expected patterns may be obscured because larval abundance covaries with resources (cryptic density dependence). Our study emphasizes the importance of covariation between the environment, habitat choice, and density dependence in understanding population dynamics across landscapes, and demonstrates the value of pairing lab and field studies.

Seasonal food webs with migrations: multi-season models reveal indirect species interactions in the Canadian Arctic tundra

Models incorporating seasonality are necessary to fully assess the impact of global warming on Arctic communities. Seasonal migrations are a key component of Arctic food webs that still elude current theories predicting a single community equilibrium. We develop a multi-season model of predator-prey dynamics using a hybrid dynamical systems framework applied to a simplified tundra food web (lemming-fox-goose-owl). Hybrid systems models can accommodate multiple equilibria, which is a basic requirement for modelling food webs whose topology changes with season. We demonstrate that our model can generate multi-annual cycling in lemming dynamics, solely from a combined effect of seasonality and state-dependent behaviour. We compare our multi-season model to a static model of the predator-prey community dynamics and study the interactions between species. Interestingly, including seasonality reveals indirect interactions between migrants and residents not captured by the static model. Further, we find that the direction and magnitude of interactions between two species are not necessarily accurate using only summer time-series. Our study demonstrates the need for the development of multi-season models and provides the tools to analyse them. Integrating seasonality in food web modelling is a vital step to improve predictions about the impacts of climate change on ecosystem functioning. This article is part of the theme issue 'The changing Arctic Ocean: consequences for biological communities, biogeochemical processes and ecosystem functioning'.

Dynamical Ising model of spatially coupled ecological oscillators

Long-range synchrony from short-range interactions is a familiar pattern in biological and physical systems, many of which share a common set of 'universal' properties at the point of synchronization. Common biological systems of coupled oscillators have been shown to be members of the Ising universality class, meaning that the very simple Ising model replicates certain spatial statistics of these systems at stationarity. This observation is useful because it reveals which aspects of spatial pattern arise independently of the details governing local dynamics, resulting in both deeper understanding of and a simpler baseline model for biological synchrony. However, in many situations a system's dynamics are of greater interest than their static spatial properties. Here, we ask whether a dynamical Ising model can replicate universal and non-universal features of ecological systems, using noisy coupled metapopulation models with two-cycle dynamics as a case study. The standard Ising model makes unrealistic dynamical predictions, but the Ising model with memory corrects this by using an additional parameter to reflect the tendency for local dynamics to maintain their phase of oscillation. By fitting the two parameters of the Ising model with memory to simulated ecological dynamics, we assess the correspondence between the Ising and ecological models in several of their features (location of the critical boundary in parameter space between synchronous and asynchronous dynamics, probability of local phase changes and ability to predict future dynamics). We find that the Ising model with memory is reasonably good at representing these properties of ecological metapopulations. The correspondence between these models creates the potential for the simple and well-known Ising class of models to become a valuable tool for understanding complex biological systems.

Numerical response of a mammalian specialist predator to multiple prey dynamics in Mediterranean farmlands

The study of rodent population cycles has greatly contributed, both theoretically and empirically, to our understanding of the circumstances under which predator-prey interactions destabilize populations. According to the specialist predator hypothesis, reciprocal interactions between voles and small predators that specialize on voles, such as weasels, can cause multiannual cycles. A fundamental feature of classical weasel-vole models is a long time-lag in the numerical response of the predator to variations in prey abundance: weasel abundance increases with that of voles and peaks approximately 1 yr later. We investigated the numerical response of the common weasel (Mustela nivalis) to fluctuating abundances of common voles (Microtus arvalis) in recently colonized agrosteppes of Castilla-y-Léon, northwestern Spain, at the southern limit of the species' range. Populations of both weasels and voles exhibited multiannual cycles with a 3-yr period. Weasels responded quickly and numerically to changes in common-vole abundance, with a time lag between prey and weasel abundance that did not exceed 4 months and occurred during the breeding season, reflecting the quick conversion of prey into predator offspring and/or immigration to sites with high vole populations. We found no evidence of a sustained, high weasel abundance following vole abundance peaks. Weasel population growth rates showed spatial synchrony across study sites approximately 60 km apart. Weasel dynamics were more synchronized with that of common voles than with other prey species (mice or shrews). However, asynchrony within, as well as among sites, in the abundance of voles and alternative prey suggests that weasel mobility could allow them to avoid starvation during low-vole phases, precluding the emergence of prolonged time lag in the numerical response to voles. Our observations are inconsistent with the specialist predator hypothesis as currently formulated, and suggest that weasels might follow rather than cause the vole cycles in northwestern Spain. The reliance of a specialized predator on a functional group of prey such as small rodents does not necessarily lead to a long delay in the numerical response by the predator, depending on the spatial and interspecific synchrony in prey dynamics.

Population cycles: generalities, exceptions and remaining mysteries

Population cycles are one of nature's great mysteries. For almost a hundred years, innumerable studies have probed the causes of cyclic dynamics in snowshoe hares, voles and lemmings, forest Lepidoptera and grouse. Even though cyclic species have very different life histories, similarities in mechanisms related to their dynamics are apparent. In addition to high reproductive rates and density-related mortality from predators, pathogens or parasitoids, other characteristics include transgenerational reduced reproduction and dispersal with increasing-peak densities, and genetic similarity among populations. Experiments to stop cyclic dynamics and comparisons of cyclic and noncyclic populations provide some understanding but both reproduction and mortality must be considered. What determines variation in amplitude and periodicity of population outbreaks remains a mystery.

Top-down limitation of lemmings revealed by experimental reduction of predators

It is generally recognized that delayed density-dependence is responsible for cyclic population dynamics. However, it is still uncertain whether a single factor can explain why some rodent populations fluctuate according to a 3-4 yr periodicity. There is increasing evidence that predation may play a role in lemming population cycles, although this effect may vary seasonally. To address this issue, we conducted an experiment where we built a large exclosure (9 ha) to protect brown lemmings (Lemmus trimucronatus) from avian and terrestrial predators. We tested the hypothesis that predation is a limiting factor for lemmings by measuring the demographic consequences of a predator reduction during the growth and peak phases of the cycle. We assessed summer (capture-mark-recapture methods) and winter (winter nest sampling) lemming demography on two grids located on Bylot Island, Nunavut, Canada from 2008 to 2015. The predator exclosure became fully effective in July 2013, allowing us to compare demography between the control and experimental grids before and during the treatment. Lemming abundance, survival and proportion of juveniles were similar between the two grids before the treatment. During the predator-reduction period, summer densities were on average 1.9× higher inside the experimental grid than the control and this effect was greatest for adult females and juveniles (densities 2.4× and 3.4× higher, respectively). Summer survival was 1.6× higher on the experimental grid than the control whereas body mass and proportion of juveniles were also slightly higher. Winter nest densities remained high inside the predator reduction grid following high summer abundance, but declined on the control grid. These results confirm that predation limits lemming population growth during the summer due to its negative impact on survival. However, it is possible that in winter, predation may interact with other factors affecting reproduction and ultimately population cycles.

Individual variation in behavioural responsiveness to humans leads to differences in breeding success and long-term population phenotypic changes

Whether human disturbance can lead to directional selection and phenotypic change in behaviour in species with limited behavioural plasticity is poorly understood in wild animal populations. Using a 19-year study on Montagu's harrier, we report a long-term increase in boldness towards humans during nest visits. The probability of females fleeing or being passive during nest visits decreased, while defence intensity steadily increased over the study period. These behavioural responses towards humans were significantly repeatable. The phenotypic composition of the breeding population changed throughout the study period (4-5 harrier generations), with a gradual disappearance of shy individuals, leading to a greater proportion of bolder ones and a more behaviourally homogeneous population. We further show that nest visit frequency increased nest failure probability and reduced productivity of shy females, but not of bold ones. Long-term research or conservation programmes needing nest visits can therefore lead to subtle but relevant population compositional changes that require further attention.

Mammalian cycles: internally defined periods and interaction-driven amplitudes

The cause of mammalian cycles—the rise and fall of populations over a predictable period of time—has remained controversial since these patterns were first observed over a century ago. In spite of extensive work on observable mammalian cycles, the field has remained divided upon what the true cause is, with a majority of opinions attributing it to either predation or to intra-species mechanisms. Here we unite the eigenperiod hypothesis, which describes an internal, maternal effect-based mechanism to explain the cycles’ periods with a recent generalization explaining the amplitude of snowshoe hare cycles in northwestern North America based on initial predator abundance. By explaining the period and the amplitude of the cycle with separate mechanisms, a unified and consistent view of the causation of cycles is reached. Based on our suggested theory, we forecast the next snowshoe hare cycle (predicted peak in 2016) to be of extraordinarily low amplitude.

Seasonal demography of a cyclic lemming population in the Canadian Arctic

1. The causes of cyclical fluctuations in animal populations remain a controversial topic in ecology. Food limitation and predation are two leading hypotheses to explain small mammal population dynamics in northern environments. We documented the seasonal timing of the decline phases and demographic parameters (survival and reproduction) associated with population changes in lemmings, allowing us to evaluate some predictions from these two hypotheses. 2. We studied the demography of brown lemmings (Lemmus trimucronatus), a species showing 3- to 4-year population cycles in the Canadian Arctic, by combining capture-mark-recapture analysis of summer live-trapping with monitoring of winter nests over a 10-year period. We also examined the effects of some weather variables on survival. 3. We found that population declines after a peak occurred between the summer and winter period and not during the winter. During the summer, population growth was driven by change in survival, but not in fecundity or proportion of juveniles, whereas in winter population growth was driven by changes in late summer and winter reproduction. 4. We did not find evidence for direct density dependence on summer demographic parameters, though our analysis was constrained by the paucity of data during the low phase. Body mass, however, was highest in peak years. 5. Weather effects were detected only in early summer when lemming survival was positively related to snow depth at the onset of melt but negatively related to rainfall. 6. Our results show that high mortality causes population declines of lemmings during summer and fall, which suggests that predation is sufficient to cause population crashes, whereas high winter fecundity is the primary factor leading to population irruptions. The positive association between snow depth and early summer survival may be due to the protective cover offered by snow against predators. It is still unclear why reproduction remains low during the low phase.

Overcompensation and phase effects in a cyclic common vole population: between first and second-order cycles

Population cycles in voles are often thought to be generated by one-year delayed density dependence on the annual population growth rate. In common voles, however, it has been suggested by Turchin (2003) that some populations exhibit first-order cycles, resulting from strong overcompensation (i.e. carrying capacity overshoots in peak years, with only an effect of the current year abundance on annual growth rates). We focus on a common vole (Microtus arvalis) population from western France that exhibits 3-year cycles. Several overcompensating nonlinear models for populations dynamics are fitted to the data, notably those of Hassell, and Maynard-Smith and Slatkin. Overcompensating direct density dependence (DD) provides a satisfactory description of winter crashes, and one-year delayed density dependence is not responsible for the crashes, thus these are not classical second-order cycles. A phase-driven modulation of direct density dependence maintains a low-phase, explaining why the cycles last three years instead of two. Our analyses suggest that some of this phase dependence can be expressed as one-year delayed DD, but phase dependence provides a better description. Hence, modelling suggests that cycles in this population are first-order cycles with a low phase after peaks, rather than fully second-order cycles. However, based on the popular log-linear second-order autoregressive model, we would conclude only that negative delayed density dependence exists. The additive structure of this model cannot show when delayed DD occurs (here, during lows rather than peaks). Our analyses thus call into question the automated use of second-order log-linear models, and suggests that more attention should be given to non-(log)linear models when studying cyclic populations. From a biological viewpoint, the fast crashes through overcompensation that we found suggest they might be caused by parasites or food rather than predators, though predators might have a role in maintaining the low phase and spatial synchrony.