Fitting individual-based models of spatial population dynamics to long-term monitoring data

Generating spatial predictions of species distribution is a central task for research and policy. Currently, correlative species distribution models (cSDMs) are among the most widely used tools for this purpose. However, a fundamental assumption of cSDMs, that species distributions are in equilibrium with their environment, is rarely fulfilled in real data and limits the applicability of cSDMs for dynamic projections. Process-based, dynamic SDMs (dSDMs) promise to overcome these limitations as they explicitly represent transient dynamics and enhance spatiotemporal transferability. Software tools for implementing dSDMs are becoming increasingly available, but their parameter estimation can be complex. Here, we test the feasibility of calibrating and validating a dSDM using long-term monitoring data of Swiss red kites (Milvus milvus). This population has shown strong increases in abundance and a progressive range expansion over the last decades, indicating a nonequilibrium situation. We construct an individual-based model using the RangeShiftR modeling platform and use Bayesian inference for model calibration. This allows the integration of heterogeneous data sources, such as parameter estimates from published literature and observational data from monitoring schemes, with a coherent assessment of parameter uncertainty. Our monitoring data encompass counts of breeding pairs at 267 sites across Switzerland over 22 years. We validate our model using a spatial-block cross-validation scheme and assess predictive performance with a rank-correlation coefficient. Our model showed very good predictive accuracy of spatial projections and represented well the observed population dynamics over the last two decades. Results suggest that reproductive success was a key factor driving the observed range expansion. According to our model, the Swiss red kite population fills large parts of its current range but has potential for further increases in density. We demonstrate the practicality of data integration and validation for dSDMs using RangeShiftR. This approach can improve predictive performance compared to cSDMs. The workflow presented here can be adopted for any population for which some prior knowledge on demographic and dispersal parameters as well as spatiotemporal observations of abundance or presence/absence are available. The fitted model provides improved quantitative insights into the ecology of a species, which can greatly aid conservation and management efforts.

Stability of patch-turnover relationships under equilibrium and nonequilibrium metapopulation dynamics driven by biogeography

Two controversial tenets of metapopulation biology are whether patch quality and the surrounding matrix are more important to turnover (colonisation and extinction) than biogeography (patch area and isolation) and whether factors governing turnover during equilibrium also dominate nonequilibrium dynamics. We tested both tenets using 18 years of surveys for two secretive wetland birds, black and Virginia rails, during (1) a period of equilibrium with stable occupancy and (2) after drought and arrival of West Nile Virus (WNV), which resulted in WNV infections in rails, increased extinction and decreased colonisation probabilities modified by WNV, nonequilibrium dynamics for both species and occupancy decline for black rails. Area (primarily) and isolation (secondarily) drove turnover during both stable and unstable metapopulation dynamics, greatly exceeding the effects of patch quality and matrix conditions. Moreover, slopes between turnover and patch characteristics changed little between equilibrium and nonequilibrium, confirming the overriding influences of biogeographic factors on turnover.

Linking monitoring and data analysis to predictions and decisions for the range-wide eastern black rail status assessment

The US Fish and Wildlife Service (USFWS) has initiated a re-envisioned approach for providing decision makers with the best available science and synthesis of that information, called the Species Status Assessment (SSA), for endangered species decision making. The SSA report is a descriptive document that provides decision makers with an assessment of the current and predicted future status of a species. These analyses support all manner of decisions under the US Endangered Species Act, such as listing, reclassification, and recovery planning. Novel scientific analysis and predictive modeling in SSAs could be an important part of rooting conservation decisions in current data and cutting edge analytical and modeling techniques. Here, we describe a novel analysis of available data to assess the current condition of eastern black rail Laterallus jamaicensis jamaicensis across its range in a dynamic occupancy analysis. We used the results of the analysis to develop a site occupancy projection model where the model parameters (initial occupancy, site persistence, colonization) were linked to environmental covariates, such as land management and land cover change (sea-level rise, development, etc.). We used the projection model to predict future status under multiple sea-level rise and habitat management scenarios. Occupancy probability and site colonization were low in all analysis units, and site persistence was also low, suggesting low resiliency and redundancy currently. Extinction probability was high for all analysis units in all simulated scenarios except one with significant effort to preserve existing habitat, suggesting low future resiliency and redundancy. With the results of these data analyses and predictive models, the USFWS concluded that protections of the Endangered Species Act were warranted for this subspecies.

A statistical forecasting approach to metapopulation viability analysis

Conservation of at-risk species is aided by reliable forecasts of the consequences of environmental change and management actions on population viability. Forecasts from conventional population viability analysis (PVA) are made using a two-step procedure in which parameters are estimated, or elicited from expert opinion, and then plugged into a stochastic population model without accounting for parameter uncertainty. Recently developed statistical PVAs differ because forecasts are made conditional on models fitted to empirical data. The statistical forecasting approach allows for uncertainty about parameters, but it has rarely been applied in metapopulation contexts where spatially explicit inference is needed about colonization and extinction dynamics and other forms of stochasticity that influence metapopulation viability. We conducted a statistical metapopulation viability analysis (MPVA) using 11 yr of data on the federally threatened Chiricahua leopard frog (Lithobates chiricahuensis) to forecast responses to landscape heterogeneity, drought, environmental stochasticity, and management. We evaluated several future environmental scenarios and pond restoration options designed to reduce extinction risk. Forecasts over a 50-yr time horizon indicated that metapopulation extinction risk was <4% for all scenarios, but uncertainty was high. Without pond restoration, extinction risk is forecasted to be 3.9% (95% CI 0-37%) by year 2066. Restoring six ponds by increasing their hydroperiod reduced extinction risk to <1% and greatly reduced uncertainty (95% CI 0-2%). Our results suggest that managers can mitigate the impacts of drought and environmental stochasticity on metapopulation viability by maintaining ponds that hold water throughout the year and keeping them free of invasive predators. Our study illustrates the utility of the spatially explicit statistical forecasting approach to MPVA in conservation planning efforts.

The rescue effect and inference from isolation-extinction relationships

The rescue effect in metapopulations hypothesises that less isolated patches are unlikely to go extinct because recolonisation may occur between breeding seasons ('recolonisation rescue'), or immigrants may sufficiently bolster population size to prevent extinction altogether ('demographic rescue'). These mechanisms have rarely been demonstrated directly, and most evidence of the rescue effect is from relationships between isolation and extinction. We determined the frequency of recolonisation rescue for metapopulations of black rails (Laterallus jamaicensis) and Virginia rails (Rallus limicola) from occupancy surveys conducted during and between breeding seasons, and assessed the reliability of inferences about the occurrence of rescue drawn from isolation-extinction relationships, including autologistic isolation measures that corrected for unsurveyed patches and imperfect detection. Recolonisation rescue occurred at expected rates, but was elevated during periods of disturbance that resulted in non-equilibrium metapopulation dynamics. Inferences from extinction-isolation relationships were unreliable, particularly for autologistic measures and for the more vagile Virginia rail.

From individual movement behaviour to landscape-scale invasion dynamics and management: a case study of lionfish metapopulations

Modelling the dynamics of small, interconnected populations, or metapopulations, can help pinpoint habitat patches that are critical for population persistence in patchy habitats. For conservation purposes, these patches are typically earmarked for protection, but for invasive species management, these patches could be targeted to hasten the populations' demise. Here, we show how metapopulation modelling, coupled with an understanding of size-dependent dispersal behaviour, can be used to help optimize the distribution of limited resources for culling specific populations of invasive Indo-Pacific lionfish (Pterois volitans) in the western Atlantic. Through simulation using fitted model parameters, we derive three insights that can inform management. First, culling lionfish from target patches reduces the probability of lionfish occupancy at surrounding patches. Second, this effect depends on patch size and connectivity, but is strongest at the local scale and decays with distance. Finally, size-dependent dispersal in lionfish means that size-selective culling can change both a population's size distribution and dispersal potential, with cascading effects on network connectivity, population dynamics and management outcomes. By explicitly considering seascape structure and movement behaviour when allocating effort to the management of invasive species, managers can optimize resource use to improve management outcomes. This article is part of the theme issue 'Linking behaviour to dynamics of populations and communities: application of novel approaches in behavioural ecology to conservation'.

Use of stochastic patch occupancy models in the California red-legged frog for Bayesian inference regarding past events and future persistence

Assessing causes of population decline is critically important to management of threatened species. Stochastic patch occupancy models (SPOMs) are popular tools for examining spatial and temporal dynamics of populations when presence-absence data in multiple habitat patches are available. We developed a Bayesian Markov chain method that extends existing SPOMs by focusing on past environmental changes that may have altered occupancy patterns prior to the beginning of data collection. Using occupancy data from 3 creeks, we applied the method to assess 2 hypothesized causes of population decline-in situ die-off and residual impact of past source population loss-in the California red-legged frog. Despite having no data for the 20-30 years between the hypothetical event leading to population decline and the first data collected, we were able to discriminate among hypotheses, finding evidence that in situ die-off increased in 2 of the creeks. Although the creeks had comparable numbers of occupied segments, owing to different extinction-colonization dynamics, our model predicted an 8-fold difference in persistence probabilities of their populations to 2030. Adding a source population led to a greater predicted persistence probability than did decreasing the in situ die-off, emphasizing that reversing the deleterious impacts of a disturbance may not be the most efficient management strategy. We expect our method will be useful for studying dynamics and evaluating management strategies of many species.

Drought-mediated extinction of an arid-land amphibian: insights from a spatially explicit dynamic occupancy model

Understanding how natural and anthropogenic processes affect population dynamics of species with patchy distributions is critical to predicting their responses to environmental changes. Despite considerable evidence that demographic rates and dispersal patterns vary temporally in response to an array of biotic and abiotic processes, few applications of metapopulation theory have sought to explore factors that explain spatiotemporal variation in extinction or colonization rates. To facilitate exploring these factors, we extended a spatially explicit model of metapopulation dynamics to create a framework that requires only binary presence-absence data, makes few assumptions about the dispersal process, and accounts for imperfect detection. We apply this framework to 22 yr of biannual survey data for lowland leopard frogs, Lithobates yavapaiensis, an amphibian that inhabits arid stream systems in the southwestern United States and northern Mexico. Our results highlight the importance of accounting for factors that govern temporal variation in transition probabilities, as both extinction and colonization rates varied with hydrologic conditions. Specifically, local extinctions were more frequent during drought periods, particularly at sites without reliable surface water. Colonization rates increased when larval and dispersal periods were wetter than normal, which increased the probability that potential emigrants metamorphosed and reached neighboring sites. Extirpation of frogs from all sites in one watershed during a period of severe drought demonstrated the influence of site-level features, as frogs persisted only in areas where most sites held water consistently and where the amount of sediment deposited from high-elevation wildfires was low. Application of our model provided novel insights into how climate-related processes affected the distribution and population dynamics of an arid-land amphibian. The approach we describe has application to a wide array of species that inhabit patchy environments, can improve our understanding of factors that govern metapopulation dynamics, and can inform strategies for conservation of imperiled species.

The role of spatial structure in the collapse of regional metapopulations

Many wildlife populations are either naturally, or as a result of human land use, patchily distributed in space. The degree of fragmentation-specifically the remaining patch sizes and habitat configuration-is an important part of population dynamics. Demographic stochasticity is also likely to play an important role in patchy habitats that host small local populations. We develop a simulation model to evaluate the significance of demographic stochasticity and the role fragmentation plays in the determination of population dynamics and the risk of extinction of populations on habitat patches. Our model is formulated as a Markov-chain stochastic process on a finite, spatially explicit array of patches in which probability of successful dispersal is a function of interpatch distance. Unlike past work, we explicitly model local population dynamics and examine how these scale up to the entire population. As a test case, we apply the model to the American pika (Ochotona princeps) population living on the ore dumps in the ghost mining town of Bodie, California. This population has been studied nearly continuously for over four decades and has been of conservation concern as the southern half of the population declined precipitously beginning in 1989. Our model suggests that both the specific configuration of habitat and landscape heterogeneity are necessary and sufficient predictors of the eventual extinction of the southern constellation of patches. This example has important implications, as it suggests that fragmentation alone can lead to regional extinctions within metapopulations.

Quantifying climate sensitivity and climate-driven change in North American amphibian communities

Changing climate will impact species' ranges only when environmental variability directly impacts the demography of local populations. However, measurement of demographic responses to climate change has largely been limited to single species and locations. Here we show that amphibian communities are responsive to climatic variability, using >500,000 time-series observations for 81 species across 86 North American study areas. The effect of climate on local colonization and persistence probabilities varies among eco-regions and depends on local climate, species life-histories, and taxonomic classification. We found that local species richness is most sensitive to changes in water availability during breeding and changes in winter conditions. Based on the relationships we measure, recent changes in climate cannot explain why local species richness of North American amphibians has rapidly declined. However, changing climate does explain why some populations are declining faster than others. Our results provide important insights into how amphibians respond to climate and a general framework for measuring climate impacts on species richness.

Increasing connectivity between metapopulation ecology and landscape ecology

Metapopulation ecology and landscape ecology aim to understand how spatial structure influences ecological processes, yet these disciplines address the problem using fundamentally different modeling approaches. Metapopulation models describe how the spatial distribution of patches affects colonization and extinction, but often do not account for the heterogeneity in the landscape between patches. Models in landscape ecology use detailed descriptions of landscape structure, but often without considering colonization and extinction dynamics. We present a novel spatially explicit modeling framework for narrowing the divide between these disciplines to advance understanding of the effects of landscape structure on metapopulation dynamics. Unlike previous efforts, this framework allows for statistical inference on landscape resistance to colonization using empirical data. We demonstrate the approach using 11 yr of data on a threatened amphibian in a desert ecosystem. Occupancy data for Lithobates chiricahuensis (Chiricahua leopard frog) were collected on the Buenos Aires National Wildlife Refuge (BANWR), Arizona, USA from 2007 to 2017 following a reintroduction in 2003. Results indicated that colonization dynamics were influenced by both patch characteristics and landscape structure. Landscape resistance increased with increasing elevation and distance to the nearest streambed. Colonization rate was also influenced by patch quality, with semi-permanent and permanent ponds contributing substantially more to the colonization of neighboring ponds relative to intermittent ponds. Ponds that only hold water intermittently also had the highest extinction rate. Our modeling framework can be widely applied to understand metapopulation dynamics in complex landscapes, particularly in systems in which the environment between habitat patches influences the colonization process.

Immigration ensures population survival in the Siberian flying squirrel

Linking dispersal to population growth remains a challenging task and is a major knowledge gap, for example, for conservation management. We studied relative roles of different demographic rates behind population growth in Siberian flying squirrels in two nest-box breeding populations in western Finland. Adults and offspring were captured and individually identifiable. We constructed an integrated population model, which estimated all relevant annual demographic rates (birth, local [apparent] survival, and immigration) as well as population growth rates. One population (studied 2002-2014) fluctuated around a steady-state equilibrium, whereas the other (studied 1995-2014) showed a numerical decline. Immigration was the demographic rate which showed clear correlations to annual population growth rates in both populations. Population growth rate was density dependent in both populations. None of the demographic rates nor the population growth rate correlated across the two study populations, despite their proximity suggesting that factors regulating the dynamics are determined locally. We conclude that flying squirrels may persist in a network of uncoupled subpopulations, where movement between subpopulations is of critical importance. Our study supports the view that dispersal has the key role in population survival of a small forest rodent.

The trajectory of dispersal research in conservation biology. Systematic review

Dispersal knowledge is essential for conservation management, and demand is growing. But are we accumulating dispersal knowledge at a pace that can meet the demand? To answer this question we tested for changes in dispersal data collection and use over time. Our systematic review of 655 conservation-related publications compared five topics: climate change, habitat restoration, population viability analysis, land planning (systematic conservation planning) and invasive species. We analysed temporal changes in the: (i) questions asked by dispersal-related research; (ii) methods used to study dispersal; (iii) the quality of dispersal data; (iv) extent that dispersal knowledge is lacking, and; (v) likely consequences of limited dispersal knowledge. Research questions have changed little over time; the same problems examined in the 1990s are still being addressed. The most common methods used to study dispersal were occupancy data, expert opinion and modelling, which often provided indirect, low quality information about dispersal. Although use of genetics for estimating dispersal has increased, new ecological and genetic methods for measuring dispersal are not yet widely adopted. Almost half of the papers identified knowledge gaps related to dispersal. Limited dispersal knowledge often made it impossible to discover ecological processes or compromised conservation outcomes. The quality of dispersal data used in climate change research has increased since the 1990s. In comparison, restoration ecology inadequately addresses large-scale process, whilst the gap between knowledge accumulation and growth in applications may be increasing in land planning. To overcome apparent stagnation in collection and use of dispersal knowledge, researchers need to: (i) improve the quality of available data using new approaches; (ii) understand the complementarities of different methods and; (iii) define the value of different kinds of dispersal information for supporting management decisions. Ambitious, multi-disciplinary research programs studying many species are critical for advancing dispersal research.

Accounting for false positive detection error induced by transient individuals

Context In metapopulations, colonisation is the result of dispersal from neighbouring occupied patches, typically juveniles dispersing from natal to breeding sites. When occupancy dynamics are dispersal driven, occupancy should refer to the presence of established, breeding populations. The detection of transient individuals at sites that are, by definition, unoccupied (i.e. false positive detections), may result in misleading conclusions about metapopulation dynamics. Until recently, the issue of false positives has been considered negligible and current efforts to account for such error have been restricted to the context of species misidentification. However, the detection of transient individuals visiting multiple sites while dispersing is a distinct source of false positives that can bias estimates of occupancy because visited sites do not contribute to metapopulation dynamics in the same way as do sites occupied by established, reproducing populations. Although transient-induced false positive error presents a challenge to occupancy studies aiming to account for all sources of detection error and estimate occupancy without bias, accounting for it has received little attention. Aims Using a novel application of an existing occupancy model, we sought to account for false positives that result from transient individuals being observed at truly unoccupied sites (i.e. where no establishment has occurred). Methods We applied a Bayesian multi-season occupancy model correcting for false negative and false positive errors, to 3 years of detection or non-detection data from a metapopulation of water voles, Arvicola amphibious, in which both types of patch-state misclassification are suspected. Key results We provide evidence that transient individuals can cause false positive detection errors. We then demonstrate the flexibility of the occupancy model to account for both false negative and false positive detection errors beyond the typical application to species misidentification. Accounting for both types of observation error reduces the bias in estimates of occupancy and avoids misleading conclusions about the status of (meta) populations by allowing for the distinction to be made between resident and transient occupancy. Conclusion In many species, transience may result in patch-state misclassification which needs to be accounted for so as to draw correct inference about metapopulation status. Making the distinction between occupancy by established populations and visitation by transients will influence how we interpret patch occupancy dynamics, with important implications for the management of wildlife. Implications The ability to estimate occupancy free of bias induced by false positive detections can help ensure that downward trends in occupancy are detected despite such declines being accompanied by increasing frequency of transients associated with, for example, reductions in mate availability or failure to establish. Our approach can be applied to any occupancy study in which false positive detections are suspected because of the behaviour of the focal species.

Why are metapopulations so rare?

Roughly 40 years after its introduction, the metapopulation concept is central to population ecology. The notion that local populations and their dynamics may be coupled by dispersal is without any doubt of great importance for our understanding of population-level processes. A metapopulation describes a set of subpopulations linked by (rare) dispersal events in a dynamic equilibrium of extinctions and recolonizations. In the large body of literature that has accumulated, the term "metapopulation" is often used in a very broad sense; most of the time it simply implies spatial heterogeneity. A number of reviews have recently addressed this problem and have pointed out that, despite the large and still growing popularity of the metapopulation concept, there are only very few empirical examples that conform with the strict classical metapopulation (CM) definition. In order to understand this discrepancy between theory and observation, we use an individual-based modeling approach that allows us to pinpoint the environmental conditions and the life-history attributes required for the emergence of a CM structure. We find that CM dynamics are restricted to a specific parameter range at the border between spatially structured but completely occupied and globally extinct populations. Considering general life-history attributes, our simulations suggest that CMs are more likely to occur in arthropod species than in (large) vertebrates. Since the specific type of spatial population structure determines conservation concepts, our findings have important implications for conservation biology. Our model suggests that most spatially structured populations are panmictic, patchy, or of mainland-island type, which makes efforts spent on increasing connectivity (e.g., corridors) questionable. If one does observe a true CM structure, this means that the focal metapopulation is on the brink of extinction and that drastic conservation measures are needed.

Impact of cattle grazing on the occupancy of a cryptic, threatened rail

Impacts of livestock grazing in arid and semiarid environments are often concentrated in and around wetlands where animals congregate for water, cooler temperatures, and green forage. We assessed the impacts of winter-spring (November-May) cattle grazing on marsh vegetation cover and occupancy of a highly secretive marsh bird that relies on dense vegetation cover, the California Black Rail (Laterallus jamaicensis coturniculus), in the northern Sierra Nevada foothills of California, U.S.A. Using detection-nondetection data collected during repeated call playback surveys at grazed vs. ungrazed marshes and a "random changes in occupancy" parameterization of a multi-season occupancy model, we examined relationships between occupancy and habitat covariates, while accounting for imperfect detection. Marsh vegetation cover was significantly lower at grazed marshes than at ungrazed marshes during the grazing season in 2007 but not in 2008. Winter-spring grazing had little effect on Black Rail occupancy at irrigated marshes. However, at nonirrigated marshes fed by natural springs and streams, grazed sites had lower occupancy than ungrazed sites. Black Rail occupancy was positively associated with marsh area, irrigation as a water source, and summer vegetation cover, and negatively associated with marsh isolation. Residual dry matter (RDM), a commonly used metric of grazing intensity, was significantly associated with summer marsh vegetation cover at grazed sites but not spring cover. Direct monitoring of marsh vegetation cover, particularly at natural spring- or stream-fed marshes, is recommended to prevent negative impacts to rails from overgrazing.

Estimating population impacts via dynamic occupancy analysis of Before-After Control-Impact studies

Estimating environmental impacts on populations is one of the main goals of wildlife monitoring programs, which are often conducted in conjunction with management actions or following natural disturbances. In this study we investigate the statistical power of dynamic occupancy models to detect changes in local survival and colonization from detection-nondetection data, while accounting for imperfect detection probability, in a Before-After Control-Impact (BACI) framework. We simulated impacts on local survival and/or detection probabilities, and asked questions related to: (1) costs and benefits of different analysis models, (2) confounding changes in detection with changes in local survival, (3) sampling design trade-offs, and (4) species with low vs. high rates of turnover. Estimating seasonal effects on local survival and colonization, as opposed to estimating Before-After effects, had little effect on the power to detect changes in local survival. Estimating a parameter that accounted for pretreatment differences in local survival between Control and Impact sites decreased power by 50%, but it was critical to include when such differences existed. When the experimental treatment had a negative impact on species detectability but analysis assumed constant detection, the Type I error rates were dramatically inflated (0.20 0.33). In general, there was low power (< 0.5) to detect a 50% decrease in local survival for all combinations of sites (N = 50 vs. 100), seasons sampled (8 vs. 12), and visits per site per season (4 vs. 6). Unbalanced designs performed worse than balanced designs, with the exception of the case of treatments being implemented in different seasons at different sites. Adding more control sites improved the ability to detect changes in local survival. Surveying more seasons after impact resulted in modest power gains, but at least three seasons before impact were required to successfully implement BACI occupancy studies. Turnover rates had a low impact on power. Occupancy studies conducted in a BACI design offer the opportunity to detect environmental impacts on wildlife populations without the costs of intensive studies. However, given the low power to detect small changes (20%) in local survival, these studies should be used when researchers are confident that major treatment impacts will occur or very large sample sizes are obtainable.