By the same authors

From the same journal

Process from pattern in the distribution of an endangered leaf beetle

Research output: Contribution to journalArticle

Author(s)

Department/unit(s)

Publication details

JournalEcography
DatePublished - Apr 2009
Issue number2
Volume32
Number of pages10
Pages (from-to)259-268
Original languageEnglish

Abstract

We investigated whether signals of known dispersal processes and habitat patch turnover could be detected in a snapshot of the distribution of the tansy leaf beetle Chrysolina graminis among patches of its host plant tansy Tanacetum vulgare. Beetle occupancy in 1305 patches was analysed using autologistic generalised additive models (GAMs). These model spatial autocorrelation with an autocovariate calculated as the distance-weighted rate of occupancy among neighbouring patches. The autocovariate that best explained beetle occupancy was one which represented the active search for patches during beetle dispersal, included a distance weight that closely matched a previously fitted dispersal kernel and had neighbourhood sizes encompassing similar to 95% of known dispersal distances. Autocovariates distinguishing between neighbours on the same and opposite riverbanks outperformed those that did not, revealing the river as a barrier to dispersal. Differentiating between up and downstream autocorrelation did not improve model fit, as is consistent with the beetle's lack of directional bias in dispersal. Habitat connectivity (the extent to which it was surrounded by other patches) did not appear to affect beetle occupancy in the field, while positive effects were found for distributions simulated from the GAM. We argue that this reflects a non-equilibrium distribution driven by slow responses to high rates of habitat patch turnover due to limited dispersal ability. Our findings suggest that presence/absence snapshots can reveal patterns of dispersal and be used to test whether species' ranges are at equilibrium. Such information is important for effective conservation so the possibility of inferring these patterns from distribution data is an appealing one.

    Research areas

  • SPATIAL AUTOCORRELATION, AUTOLOGISTIC REGRESSION, SPECIES DISTRIBUTION, DISTRIBUTION MODELS, DYNAMIC LANDSCAPES, ECOLOGICAL THEORY, PERSISTENCE, PREDICTION, DEPENDENCE, VEGETATION

Discover related content

Find related publications, people, projects, datasets and more using interactive charts.

View graph of relations