Using monitoring data gathered by volunteers to predict the potential distribution of the invasive alien bumblebee Bombus terrestris

We explored the use of ecological niche-modeling from data compiled in a participatory program of thousands of volunteer observers to quantitatively evaluate the spatial risk of invasion by the alien species Bombus terrestris. B. terrestris has been commercially introduced to Japan for the pollination of greenhouse tomato crops since 1991. Recently, naturalized populations of this species have spread rapidly, particularly in Hokkaido, northern Japan. Two niche-based models were constructed from independent data sets (presence-absence and presence-only data sets) and were used to validate one another. Both models performed well and indicated that the data compiled by participatory programs were useful for generating predictive models of the potential distribution of this invasive species. The models also revealed that the potential distribution of B. terrestris is negatively related to the proportion of woodland area, which corresponds to the known biology of this species. In contrast, the effect of the number of colonies introduced for tomato pollination (represented by tomato production) did not significantly affect the distribution of B. terrestris, indicating that its spread is now entering a phase of dispersal from established feral populations and that primary dispersal from commercial colonies in greenhouses is no longer a limiting factor for the distribution of this invasive species.     

基于Logistic-GTWR模型的武汉城市圈城镇用地扩展驱动力分析

该文以武汉城市圈为例,利用1995年、2000年、2005年、2010年、2015年5期土地利用数据、社会经济因子、邻域因子和自然因子,构建耦合空间异质性和时间非平稳性的Logistic-GTWR(逻辑斯蒂-时空地理加权回归)模型,挖掘武汉城市圈城镇用地扩展驱动力及其时空分异格局。结果表明:1)耦合空间异质性和时间非平稳性的Logistic-GTWR模型比全局Logistic回归模型和Logistic-GWR(逻辑斯蒂-地理加权回归)模型具有更好的拟合效果,验证了Logistic-GTWR模型在城镇用地扩展驱动力分析研究中的适用性。2)武汉城市圈城镇用地扩展的不同影响因子系数时空分异格局各不相同,人口因子高值分布区围绕武汉市先收缩后扩散并逐渐稳定,最终形成绕武汉城市圈中部沿线周边环状分布格局;地均GDP因子高值分布区先向西部转移,后逐渐趋于均衡,最终形成绕武汉、麻城和崇阳县周边集聚分布的空间格局。武汉市极化作用占据主导地位,且影响范围不断扩大。3)人口和经济是20年间武汉城市圈城镇扩展的核心驱动因子,且影响力呈逐年上升趋势;国道、高速和省道影响力整体呈略微升高态势。4)武汉城市圈中部及东部城市多为人口、经济核心驱动型城市,高程影响在武汉市发展中逐渐凸显;西部城市主要受高程限制,并以交通及经济驱动为主,东西差异显著;但西部城市1995年以来人口和经济的作用强度逐步增加。人口因子作用强度在咸宁和黄冈呈逐渐上升趋势,经济因子作用强度在武汉、孝感、仙桃、天门、潜江和黄冈逐年上升。     

Simulating the effects of different disturbance regimes on Cortaderia selloana invasion

Disturbances favour plant invasions because they provide a pulse of resources for seedling establishment and growth. However, little attention has been paid to the importance of the different disturbance regimes in the extent of invasion. In this paper we asked to what extent invasion depends on spatial and temporal patterns of disturbance. By using the spatially explicit model Fateland, we simulated how Cortaderia selloana, a large-wind dispersed perennial grass invading Mediterranean grasslands, responds to different medium-term (up to 55 years) disturbance regimes. The different scenarios combine disturbance frequency, extent of disturbance and temporal and spatial disturbance variability, and they roughly mimic real situations of habitat management and disturbances in cultural landscapes.The model predicted that the proportion of invaded area would increase with disturbance frequency. The extent of the disturbance overcame the importance of frequency. Aggregated disturbances randomly distributed in space had a more positive effect on invasion than scattered disturbances at repeatedly fixed locations. Whether the disturbances regularly occurred in time or were irregularly distributed did not influence the percentage of invaded landscape. Our simulation analysis unambiguously supports the hypothesis that disturbances are essential for the persistence and expansion of Cortaderia sellona. The high amount of seed rain from a few individuals (e.g., gardens) make this species very sensitive to landscape changes produced by different disturbance patterns (e.g., land abandonment, habitat management or rubbish dumping). Our results can contribute to redirect land management options.     

Crofting and bumblebee conservation: The impact of land management practices on bumblebee populations in northwest Scotland

The northwest of Scotland is a stronghold for two of the UK’s rarest bumblebee species, Bombus distinguendus and Bombus muscorum. The predominant form of agricultural land management in this region is crofting, a system specific to Scotland in which small agricultural units (crofts) operate rotational cropping and grazing regimes. Crofting is considered to be beneficial to a wide range of flora and fauna. However, currently there is a lack of quantitative evidence to support this view with regard to bumblebee populations. In this study we assessed the effect of land management on the abundance of foraging bumblebees and the availability of bumblebee forage plants across crofts in northwest Scotland. The results of our study show that current crofting practices do not support high densities of foraging bumblebees. Traditional crofting practice was to move livestock to uplands in the summer, but this has been largely abandoned. Summer sheep grazing of lowland pasture had a strong negative impact on bumblebee abundance and forage plant availability throughout the survey period. The use of specific ‘bird and bee’ conservation seed mixes appears to improve forage availability within the crofted landscape, although the number of bees observed remained low. Of the forage plants available, the three most frequently visited species were from the Fabaceae. We therefore conclude that the creation of agri-environment schemes which promote the use of Fabaceae-rich seed mixes and encourage the removal of sheep grazing on lowland areas throughout the summer are essential in order to conserve bumblebee populations within crofted areas.     

Pollination of the invasive exotic shrub Lupinus arboreus (Fabaceae) by introduced bees in Tasmania

Exotic plant invasions threaten ecological communities world-wide. Some species are limited by a lack of suitable pollinators, but the introduction of exotic pollinators can facilitate rapid spread. In Tasmania, where many non-native plants are naturalised, exotic honeybees (Apis mellifera) and bumblebees (Bombus terrestris) have become established. We determined how these species affect the pollination of Lupinus arboreus, an invasive, nitrogen-fixing shrub, which is rarely visited by native pollinators. The proportion of flowers setting seed and the number of ovules fertilised per flower were positively related to the visitation rates of both exotic bee species. There was no effect of bee visitation rates on the proportion of seeds aborted prior to maturity, possibly due to post-fertilisation environmental constraints. We conclude that the spread of B. terrestris may not alter the fecundity of L. arboreus because of the pollination service provided by A. mellifera, and discuss potential interactions between these two bee species.     

Public and professional views on invasive non-native species – A qualitative social scientific investigation

The potential of invasive non-native species to modify ecosystems is well established. Yet, the resulting visibility of the field of invasion biology and its growing influence on policy decisions has led to the development of persistent critique. In addition, public opposition to the removal of non-native species has repeatedly delayed interventions, sometimes to the point where eradication has become impossible. Understanding the views of the lay-public is thus of crucial importance. Our study investigated views on invasive non-native species as held by both the general public and ecologists. We chose an in-depth qualitative social scientific approach to identify species attributes that people draw on to evaluate biological invasions and potential management options. Interviews and focus group discussions (n = 79 participants) were conducted in the North-east of Scotland, and included members of the public from a wide range of backgrounds, as well as research ecologists and professionals dealing with invasive non-native species. Two key arguments, namely (i) harmfulness of a species and (ii) human responsibility for its spread, shaped both ecologists’ and lay-people’s discussions. Non-nativeness was understood by our participants in various ways and did not necessarily raise concerns about a species. Our results indicate that for information campaigns on species management to be effective, values inherent to the debate about invasive non-native species need addressing in a more explicit and transparent manner. Arguments such as human responsibility that are currently neglected in the scientific discourse, but that were important criteria for a wide range of participants, should be made explicit.     

Development and evaluation of predictive habitat models to assist the conservation planning of a threatened lucanid beetle, Hoplogonus simsoni, in north-east Tasmania

The use of predictive habitat distribution models by land managers in the conservation management of threatened species is increasing. Few models, however, are subsequently field-checked and evaluated. This study evaluates the statistical strength and usefulness for conservation purposes of three predictive habitat models developed for a threatened stag beetle, Hoplogonus simsoni, found in the wet eucalypt forests and mixed/rainforests of north-east Tasmania. The relationship between various environmental variables for which spatial (GIS) information was available and the density, frequency of occurrence and presence/absence of the species was investigated using generalised linear modelling. Models developed were coupled with the GIS data to develop maps of predicted occurrence within the species’ range, grouped into categories of habitat quality. The models found that altitude, aspect, slope, distance to nearest stream and overstorey tree height were significantly associated with the occurrence of the species. Evaluation of the statistical strength of the models with independent data of species’ occurrence collected at 95 sites found that the density model performed poorly with little correlation between predicted and observed densities of the species. The frequency of occurrence model, however, showed a moderate ability to predict both species’ abundance and presence/absence. The presence/absence model had a similar discriminatory ability in predicting presence or absence of H. simsoni, but also showed some potential as an indirect predictor of species’ abundance. Assuming a correlation between relative abundance and habitat quality, the frequency of occurrence predictive model appeared to be the better and more direct discriminator of high quality habitat relative to the other models. The value of species’ habitat models and the need to evaluate their utility in the development of conservation strategies are discussed.     

The roles of habitat features, disturbance, and distance from putative source populations in structuring alien plant invasions at the urban/wildland interface on the Cape Peninsula, South Africa

Natural areas are becoming increasingly fragmented and embedded in an urban matrix. Natural and semi-natural areas at the urban/wildland interface are threatened by a variety of ‘edge effects’, and are especially vulnerable to invasion by introduced plants, with suburban gardens acting as significant sources of alien propagules. Urban/wildland interfaces also provide access for humans, leading to various types of disturbance. Alien plant invasions are one of the biggest threats facing remaining natural areas on the Cape Peninsula, South Africa. The area provides an ideal opportunity to study the dynamics of invasions at the urban/wildland interface, since the largest natural area, the Table Mountain National Park (TMNP), is surrounded by the city of Cape Town. We explored invasion patterns in Newlands Forest (a small section of the TMNP) and detailed the roles of habitat features and distance from putative source populations in three main habitat types: natural Afromontane forest, riverine woodland habitats, and plantations of exotic pines (Pinus radiata and P. pinaster). We also examined the role of disturbance in driving invasions in two of these habitat types (Afromontane forest and pine plantations). We hypothesized that alien richness and alien stem density would decrease with distance from the urban/wildland interface, and that alien richness and alien stem density would increase with increasing levels of human disturbance.Distance from putative source populations and levels of anthropogenic disturbance influenced alien richness in Newlands Forest but not alien stem density. Alien richness decreased significantly with distance from presumed sources in the pine habitat, and increased significantly with disturbance in the forest habitat. Percentage overstorey cover and soil pH were important environmental variables associated with alien plant species. A socio-economic approach is discussed as being the most effective approach to the management and prevention of alien plant species in Newlands Forest.     

Using forest structure and composition to predict the occurrence of vertebrate species in Douglas-Fir forests of British Columbia

Forest managers require an understanding of how vertebrate species respond and persist within the dynamics of changing forest environments so that management strategies can retain and recruit structural aspects necessary for the persistence of populations. Species-habitat models are often used to understand these relationships and are subsequently used to manage landscapes. We tested several species-habitat models to predict the presence or absence of a range of vertebrate species (n = 55) and to determine the potential of using vertical and horizontal measures of forest structure as a surrogate of species occurrence. We validated models with temporally and spatially independent data. Some of the models had good predictive accuracy that was retained when validated and thus have application in terms of implementation as management tools. Modelling success varied, however, depending on whether plot or stand data were used. Many models included variables related to spatial relationships of structures. Few models were reliable when applied to independent data; therefore, our results indicate that models cannot be assumed to be applicable in different years or applied outside the area where the model was developed, even with similar spatial and temporal contexts. Overall, we did not find robust relationships necessary to guide management targets for retention and recruitment of specific forest structures. Therefore, using these habitat models as surrogates for monitoring species occurrence is limited. Monitoring aspects of habitat should still be included as part of biodiversity monitoring programs because preservation of structures known to be negatively affected by harvesting (e.g., dead wood, large trees, closed canopies, continuous forests) contributes to local and landscape heterogeneity and has been shown to affect species presence in this study and others.     

Are forests potential landscape barriers for foraging bumblebees? Landscape scale experiments with Bombus terrestris agg. and Bombus pascuorum (Hymenoptera, Apidae)

Habitat fragmentation is generally considered to have detrimental effects on insect movement and associated pollen flow between plant populations. Against this background the role of forests as potential barriers for foraging bumblebees of the species Bombus terrestris agg. and Bombus pascuorum was studied in different experiments. Bombus terrestris agg. workers were found foraging at distances of up to 2.2 km from their nests. B. terrestris agg. as well as B. pascuorum individuals crossed 600 m of forestland between floral mass resources (Phacelia tanacetifolia, Helianthus annuus), although in general a high degree of site fidelity was observed. B. pascuorum workers accepted artificial floral resources within the forest, whereas the minor use of resources below forest canopy observed for B. terrestris agg. possibly indicates a preference for direct forage flights, probably leading above the forest canopy. Our results warn against experiments with genetically modified crop species with potential bumblebee pollination (e.g. Brassica napus), in which an isolating effect of forests is assumed.     

Earthworm invasion alters enchytraeid community composition and individual biomass in northern hardwood forests of North America

European earthworms are invading many ecosystems worldwide and fundamentally transform habitats by acting as dominant ecosystem engineers. However, there is little knowledge of the consequences of earthworm invasion on the composition and diversity of native soil organisms. Particularly functionally similar groups, such as enchytraeids (Annelida: Enchytraeidae), may be affected through changes in the chemical and physical properties of the soil, but also due to competition for resources. In 2010–2011, we studied the impact of earthworm invasion on enchytraeids at two sites in the northern hardwood forests of North America: one site within the Chippewa National Forest in northern Minnesota and one site in the Chequamegon-Nicolet National Forest in northern Wisconsin, USA. At each site, three plots were sampled along a transect, representing (1) a non-invaded or very slightly invaded area, (2) the leading edge of earthworm invasion and (3) a heavily invaded area with an established population of the anecic earthworm Lumbricus terrestris (among other species). In total, 29 enchytraeid (morpho)species were identified (some yet to be formally described, several first or second records for the continent); of those 24 occurred at the Minnesota site and 17 at the Wisconsin site. The structure of enchytraeid assemblages differed significantly among the three invasion stages, although this was not equally pronounced at the two sites. Each stage was characterized by one or several indicator species. Mean enchytraeid densities (10,700–30,400 individuals/m²) did not differ significantly among the invasion stages, but were lowest at the leading edge of earthworm invasion at both sites. In the heavily invaded plot at the Minnesota site, the mean enchytraeid density and biomass in L. terrestris middens were significantly higher than in soil in-between the middens. This was due to a pronounced effect of L. terrestris middens in the uppermost 3 cm of soil. Differences in biomass among earthworm invasion stages were most apparent for mean individual biomass. This was significantly higher in the heavily invaded area than at the leading edge or in the non-invaded area at the Minnesota site. Compositional changes of the enchytraeid assemblage are likely to result in changes in the functioning of soil foods webs. Our results suggest that earthworm invasions can cause a loss of native species in soil, including heretofore unknown ones, that might go unnoticed.     

Computationally Efficient Statistical Differential Equation Modeling Using Homogenization

Statistical models using partial differential equations (PDEs) to describe dynamically evolving natural systems are appearing in the scientific literature with some regularity in recent years. Often such studies seek to characterize the dynamics of temporal or spatio-temporal phenomena such as invasive species, consumer-resource interactions, community evolution, and resource selection. Specifically, in the spatial setting, data are often available at varying spatial and temporal scales. Additionally, the necessary numerical integration of a PDE may be computationally infeasible over the spatial support of interest. We present an approach to impose computationally advantageous changes of support in statistical implementations of PDE models and demonstrate its utility through simulation using a form of PDE known as “ecological diffusion.” We also apply a statistical ecological diffusion model to a data set involving the spread of mountain pine beetle (Dendroctonus ponderosae) in Idaho, USA. This article has supplementary material online.     

Spatial distribution of spiders and epedaphic Collembola in an environmentally heterogeneous forest floor habitat

Describing the biotic and abiotic processes that are responsible for the formation of spatial patterns in predators and their prey is crucial for improving our understanding of food–web interactions. We studied the spatial distribution of four abundant spider species and three common groups of epedaphic Collembola prey in a beech-dominated (Fagus sylvatica) forest floor habitat and related the observed patterns to environmental heterogeneity, overall predator activity (all ground beetles and spiders) and prey availability (all Collembola) at the local scale. Spiders and epedaphic Collembola were sampled over 392 days in a spatially explicit design based on a regular grid of 25 pitfall traps (inter-trap distance 100 m). Environmental heterogeneity was characterized by cover of moss and litter as well as the amount of dead wood at each trap location. We first used the index of dispersion to characterize the spatial distribution of spider species and Collembola and then related the observed patterns to environmental heterogeneity, predator and prey availability while testing for spatial autocorrelation within the same models. All taxa were significantly more aggregated than expected from the assumption of random distribution. The distribution of spider species was positively (Coelotes terrestris) or negatively (Tenuiphantes zimmermanni and Tapinocyba insecta) related to the cover of moss and negatively related to litter cover (C. terrestris) or the local availability of prey (T. insecta). The distribution of Collembola was negatively related to local litter cover (Lepidocyrtus spp.) and positively related to the amount of medium deadwood pieces (all other Entomobryidae). Our study suggests that none of the spider species preferred areas of low overall predator activity density. Moreover, it does not indicate association of spider species to prey-rich areas at the analyzed scale of 100 m. It further highlights the importance of environmental heterogeneity, as different habitat properties differentially affected the local activity density of spiders and Collembola and thus considerably contributed to the understanding of distribution patterns.     

Patterns of fish invasions in the Great Plains of North America

A basic step toward controlling invasive species outbreaks is to predict which assemblages or habitats are most susceptible to invasion and effects of invading species on native communities. However, because predictive models may not work across a wide range of conditions it is important to examine processes associated with fish invasions across different regions. In this study, we use data from 949 sites in Oklahoma and Kansas to examine spatial patterns of species invasions in the Great Plains region of the US. Of these sites, 16 were intensively sampled between 1977 and 1994 and used to evaluate temporal patterns of species introductions. We found a trend of increasing introduced species richness without a decline in native species richness over 18 years. However, total abundance and community structure of native assemblages has changed at several of these sites during this time period. Because some of these changes occurred at sites with few or no introduced species, factors other than interactions with introduced fishes may have been responsible for this pattern. A stepwise multiple regression model that included human population size, native species richness, mean annual precipitation and drainage area explained 43% of the variation in the number of non-native species across all sites. Most introduced species in this region have native ranges within the Mississippi River basin and therefore are presumably adapted to regional environmental conditions. Great Plains streams have, to some extent, resisted invasion by introduced species; however, our analysis shows that proximity to dense human populations and importance to humans has largely structured species introductions and can partly explain spatial patterns of species invasions in this region.     

An assessment of the effectiveness of a large,national-scale invasive alien plant control strategy in South Africa

This paper presents an assessment of a large, national-scale alien plant control program that has operated in South Africa for 15 years. We reviewed data from three national-level estimates of the extent of invasion, records of the costs and spatial extent of invasive species control operations, assessments of the effectiveness of biological control, and smaller-scale studies. The 19 most important invasive taxa, mainly trees, in terrestrial biomes were identified. The effectiveness of control efforts on the extent of invasion of these taxa was assessed. Control costs over 15 years amounted to 3.2 billion rands (US$457 million), more than half of which was spent on 10 taxa, the most prominent being in the genera Acacia, Prosopis, Pinus and Eucalyptus. Despite substantial spending, control operations were in many cases applied to a relatively small portion of the estimated invaded area, and invasions appear to have increased, and remain a serious threat, in many biomes. Our findings suggest that South Africa’s national-scale strategy to clear invasive alien plants should be substantially modified if impacts are to be effectively mitigated. Rather than attempting to control all species, and to operate in all areas, a more focused approach is called for. This would include prioritising both the species and the areas, and setting goals and monitoring the degree to which they are achieved, within a framework of adaptive management. A greater portion of funding should also be directed towards biological control, where successes have been most notable.     

Amphibian-killing fungus loses genetic diversity as it spreads across the New World

Emergent infectious diseases are a severe threat to global biodiversity, thus conservation biologists need to understand the emergence, spread, and evolution of pathogens to identify factors driving disease outbreaks. Amphibian chytridiomycosis is a recently emerged amphibian disease caused by the fungus Batrachochytrium dendrobatidis (Bd) that has led to species extinctions and declines worldwide. The spatio-temporal dynamics of pathogen occurrence and disease outbreaks, and comparative genomic analyses of global Bd strains, support the spreading pathogen hypothesis (SPH) with pandemics occurring after introduction of Bd into naïve host populations. Here we used population genetics of the amphibian-killing fungus to test for genetic consequences of pathogen spread. Our population genetic analyses are consistent with the spread of Bd from North to Central America, based on low genetic diversity, reduction in heterozygosity, and increased allele fixation in Bd from recently infected populations. Our findings confirm the spread of Bd in the New World, and indicate that future conservation efforts should focus on: (i) functional consequences (such as changes in pathogenicity) of these genetic changes, and (ii) public education and restrictions on wildlife trade to help slow spread of the pathogen at the invasion front.     

Exotic annuals reduce soil heterogeneity in coastal sage scrub soil chemical and biological characteristics

Exotic plant invasions alter ecosystem structure and function above- and below-ground through plant–soil feedbacks. The resistance of ecosystems to invasion can be measured by the degree of change in microbial communities and soil chemical pools and fluxes, whereas their resilience can be measured by the ability to recover following restoration. Coastal sage scrub (CSS) is one of the most highly invaded ecosystems in the US but the response of CSS soils to exotic plant invasion is little known. We examined resistance and resilience of CSS soil chemical and biological characteristics following invasion of exotic annual grasses and forbs and restoration of the native plant community. We hypothesized that invasion of exotic plant species would change biological and chemical characteristics of CSS soils by altering soil nutrient inputs. Additionally, we expected that if exotic plants were controlled and native plants were restored, native soil characteristics would recover. We sampled two locations with invaded, restored and native CSS for plant community composition, soil chemistry and microbial communities, and phospholipid fatty acid (PLFA) profiles. Communities invaded by exotic annuals were resistant to some measured parameters but not others. Extractable nitrogen pools decreased, nitrogen cycling rates increased, and microbial biomass and fungal:bacterial ratios were altered in invaded soils, and these effects were mediated by the phenological stage of the dominant plant species. The largest impact of invasion on soils was an overall reduction of spatial heterogeneity in soil nutrients, nutrient cycling and microbial communities. Restored plots tended to recover in most biotic and chemical parameters including increased resource heterogeneity compared to invaded plots, suggesting that CSS soils are resilient but not resistant to invasion.     

Life history characteristics of Lumbricus terrestris and effects of the veterinary antiparasitic compounds ivermectin and fenbendazole

Earthworms in pastureland may be exposed to residues of veterinary medicines excreted in the faeces of treated cattle. Possible sub-lethal effects of these drugs on life history characteristics are critical to earthworm populations, but there are no internationally accepted guidelines for the laboratory testing that is essential as a supplement to field studies. In the laboratory, Lumbricus terrestris were kept in artificial soil and offered dung collected from cattle with sustained-release boluses of ivermectin or fenbendazole, or from untreated cattle (control). Earthworms were therefore exposed to drug residues in concentrations excreted naturally by treated cattle. Furthermore, worms were exposed to the drug in the natural way, i.e. by ingestion. A matrix model based on survival and reproductive rates as functions of life-stages was used to quantify the population dynamics of L. terrestris. There were no adverse effects of either ivermectin or fenbendazole on the individual survival and growth of L. terrestris. Mature worms produced on average 2-3 cocoons per worm per month irrespective of drug treatment, and about 50% of these cocoons hatched successfully. The median time to hatching (incubation time) depended on the age of the parent worms and was significantly lower for cocoons in the ivermectin group compared to controls, as the worms became older. The incubation time varied considerably in all groups; even within a batch of cocoons produced during 1 month by one pair of worms, the period sometimes varied by more than 1 year. The intrinsic rate of natural increase (r≈0.0044 d⁻¹) was not affected by either drug treatment. The methods and the model developed in this study are suitable for long-term studies on L. terrestris and give ecologically relevant information on population dynamics. There were no negative effects of ivermectin and fenbendazole on life history characteristics of L. terrestris and it is unlikely that earthworm populations will be affected in the field following normal use of sustained-release boluses in cattle. Finally, the matrix model points to adult survival rate and cocoon viability as the most important variables to be included in future ecotoxicological tests on L. terrestris.     

Nonparametric Threshold Model of Zero-Inflated Spatio-Temporal Data with Application to Shifts in Jellyfish Distribution

There is increasing scientific interest in studying the spatial distribution of species abundance in relation to environmental variability. Jellyfish in particular have received considerable attention in the literature and media due to regional population increases and abrupt changes in distribution. Jellyfish distribution and abundance data, like many biological datasets, are characterized by an excess of zero counts or nonstationary processes, which hampers their analyses by standard statistical methods. Here we further develop a recently proposed statistical framework, the constrained zero-inflated generalized additive model (COZIGAM), and apply it to a spatio-temporal dataset of jellyfish biomass in the Bering Sea. Our analyses indicate systematic spatial variation in the process that causes the zero inflation. Moreover, we show strong evidence of a range expansion of jellyfish from the southeastern to the northwestern portion of the survey area beginning in 1991. The proposed methodologies could be readily applied to ecological data in which zero inflation and spatio-temporal nonstationarity are suspected, such as data describing species distribution in relation to changes of climate-driven environmental variables. Some supplemental materials including an animation of jellyfish annual biomass and web appendices are available online.