Mechanistic invasive species management models and their application in conservation

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.

Priorities for improving predictions of vessel-mediated marine invasions

Nonindigenous marine species are impacting the integrity of marine ecosystems worldwide. The invasion rate is increasing, and vessel traffic, the most significant human-assisted transport pathway for marine organisms, is predicted to double by 2050. The ability to predict the transfer of marine species by international and domestic maritime traffic is needed to develop cost-effective proactive and reactive interventions that minimise introduction, establishment and spread of invasive species. However, despite several decades of research into vessel-mediated species transfers, some important knowledge gaps remain, leading to significant uncertainty in model predictions, often limiting their use in decision making and management planning. In this review, we discuss the sequential ecological process underlying human-assisted biological invasions and adapt it in a marine context. This process includes five successive stages: entrainment, transport, introduction, establishment, and the subsequent spread. We describe the factors that influence an organism's progression through these stages in the context of maritime vessel movements and identify key knowledge gaps that limit our ability to quantify the rate at which organisms successfully pass through these stages. We then highlight research priorities that will address these knowledge gaps and improve our capability to manage biosecurity risks at local, national and international scales. We identified four major data and knowledge gaps: (1) quantitative rates of entrainment of organisms by vessels; (2) the movement patterns of vessel types lacking maritime location devices; (3) quantifying the release (introduction) of organisms as a function of vessel behaviour (e.g. time spent at port); and (4) the influence of a species' life history on establishment success, for a given magnitude of propagule pressure. We discuss these four research priorities and how they can be addressed in collaboration with industry partners and stakeholders to improve our ability to predict and manage vessel-mediated biosecurity risks over the coming decades.

Partial differential equation models for invasive species spread in the presence of spatial heterogeneity

Models of invasive species spread often assume that landscapes are spatially homogeneous; thus simplifying analysis but potentially reducing accuracy. We extend a recently developed partial differential equation model for invasive conifer spread to account for spatial heterogeneity in parameter values and introduce a method to obtain key outputs (e.g. spread rates) from computational simulations. Simulations produce patterns of spatial spread which appear qualitatively similar to observed patterns in grassland ecosystems invaded by exotic conifers, validating our spatially explicit strategy. We find that incorporating spatial variation in different parameters does not significantly affect the evolution of invasions (which are characterised by a long quiescent period followed by rapid evolution towards to a constant rate of invasion) but that distributional assumptions can have a significant impact on the spread rate of invasions. Our work demonstrates that spatial variation in site-suitability or other parameters can have a significant impact on invasions and must be considered when designing models of invasive species spread.

Eradicating an invasive mammal requires local elimination and reduced reinvasion from an urban source population

Invasive mammal eradications are increasingly attempted across large, complex landscapes. Sequentially controlled management zones can be at risk of reinvasion from adjacent uncontrolled areas, and managers must weigh the relative benefits of ensuring complete elimination from a zone or minimizing reinvasion risk. This is complicated in urban areas, where habitat heterogeneity and a lack of baseline ecological knowledge increase uncertainty. We applied a spatial agent-based model to predict the reinvasion of a well-studied species, the brushtail possum (Trichosurus vulpecula), across an urban area onto a peninsula that is the site of an elimination campaign in Aotearoa New Zealand. We represented fine-scale urban habitat heterogeneity in a land cover layer and tested management scenarios that varied four factors: the density of possums remaining following an elimination attempt, the maintenance trap density on the peninsula, and effort expended toward preventing reinvasion by means of a high-density trap buffer at the peninsula isthmus or control of the source population adjacent to the peninsula. We found that achieving complete elimination on the peninsula was crucial to avoid rapid repopulation. The urban isthmus was predicted to act as a landscape barrier and restrict immigration onto the peninsula, but reliance on this barrier alone would fail to prevent repopulation. In combination, complete elimination, buffer zone, and source population control could reduce the probability of possum repopulation to near zero. Our findings support urban landscape barriers as one tool for sequential invasive mammal elimination but reaffirm that novel methods to expose residual individuals to control will be necessary to secure elimination in management zones. Work to characterize the urban ecology of many invasive mammals is still needed.

A mechanistic statistical approach to infer invasion characteristics of human-dispersed species with complex life cycle

The rising introduction of invasive species through trade networks threatens biodiversity and ecosystem services. Yet, we have a limited understanding of how transportation networks determine patterns of range expansion. This is partly because current analytical models fail to integrate the invader's life-history dynamics with heterogeneity in human-mediated dispersal patterns. And partly because classical statistical methods often fail to provide reliable estimates of model parameters due to spatial biases in the presence-only records and lack of informative demographic data. To address these gaps, we first formulate an age-structured metapopulation model that uses a probability matrix to emulate human-mediated dispersal patterns. The model reveals that an invader spreads along the shortest network path, such that the inter-patch network distances decrease with increasing traffic volume and reproductive value of hitchhikers. Next, we propose a Bayesian statistical method to estimate model parameters using presence-only data and prior demographic knowledge. To show the utility of the statistical approach, we analyze zebra mussel (Dreissena polymorpha) expansion in North America through the commercial shipping network. Our analysis underscores the importance of correcting spatial biases and leveraging priors to answer questions, such as where and when the zebra mussels were introduced and what life-history characteristics make these mollusks successful invaders.

Cost effectiveness of spread mitigation strategies for polyphagous shot hole borer Euwallacea fornicatus (Coleoptera: Curculionidae: Scolytinae)

Polyphagous shot hole borer Euwallacea fornicatus Eichhoff was detected in Western Australia in September 2021, and an eradication campaign funded by the Commonwealth government is underway. As part of contingency planning, we examined the cost effectiveness of alternative control strategies that could be used to mitigate urban forest impacts and maintain the benefits of trees to the local communities if eradication was not feasible. At the time this work was undertaken, decision-makers were concerned about the potential need to replace all urban trees susceptible to attack. We considered this strategy alongside less destructive strategies and assessed their cost effectiveness in terms of material and labor costs and the loss of ecosystem services resulting from reduced tree foliage. Using a stochastic simulation model, we found that a strategy that involved pruning necrotic limbs and treating trees biennially with systemic insecticide was almost always more cost effective than removing infested trees and replanting to resistant varieties. We estimated this strategy would cost A$55-110 million over 50 years, while tree removal would cost $105-195 million. A third strategy using a mix of chemical suppression and tree removal was also considered in light of new information about the pest's host preferences. With an estimated cost of $60-110 million, this strategy was only slightly more expensive than using chemical suppression alone and could actually lead to eradication if the host range is as narrow as recent survey data suggests.

Genomic data is missing for many highly invasive species, restricting our preparedness for escalating incursion rates

Biological invasions drive environmental change, potentially threatening native biodiversity, human health, and global economies. Population genomics is an increasingly popular tool in invasion biology, improving accuracy and providing new insights into the genetic factors that underpin invasion success compared to research based on a small number of genetic loci. We examine the extent to which population genomic resources, including reference genomes, have been used or are available for invasive species research. We find that 82% of species on the International Union for Conservation of Nature “100 Worst Invasive Alien Species” list have been studied using some form of population genetic data, but just 32% of these species have been studied using population genomic data. Further, 55% of the list’s species lack a reference genome. With incursion rates escalating globally, understanding how genome-driven processes facilitate invasion is critical, but despite a promising trend of increasing uptake, “invasion genomics” is still in its infancy. We discuss how population genomic data can enhance our understanding of biological invasion and inform proactive detection and management of invasive species, and we call for more research that specifically targets this area.

Branching structure of genealogies in spatially growing populations and its implications for population genetics inference

Spatial models where growth is limited to the population edge have been instrumental to understanding the population dynamics and the clone size distribution in growing cellular populations, such as microbial colonies and avascular tumours. A complete characterization of the coalescence process generated by spatial growth is still lacking, limiting our ability to apply classic population genetics inference to spatially growing populations. Here, we start filling this gap by investigating the statistical properties of the cell lineages generated by the two dimensional Eden model, leveraging their physical analogy with directed polymers. Our analysis provides quantitative estimates for population measurements that can easily be assessed via sequencing, such as the average number of segregating sites and the clone size distribution of a subsample of the population. Our results not only reveal remarkable features of the genealogies generated during growth, but also highlight new properties that can be misinterpreted as signs of selection if non-spatial models are inappropriately applied.