Optimising trait and source selection for explaining occurrence and abundance changes: A case study using British butterflies

Restoring a butterfly hot spot by large ungulates refaunation: the case of the Milovice military training range, Czech Republic

Background

Refaunation/rewilding by large ungulates represents a cost-efficient approach to managing natural biotopes and may be particularly useful for areas whose biodiversity depends on disturbance dynamics and is imperilled by successional changes. To study impacts of refaunation on invertebrates, we focused on butterflies inhabiting the former military training range Milovice, Czech Republic, refaunated since 2015 by a combination of Exmoor pony (“wild” horse), Tauros cattle (“aurochs”), and European wisent.

Methods

We analysed butterfly presence-absence patterns immediately after the military use termination (early 1990s), prior to the refaunation (2009), and after it (2016–19); and current abundance data gained by monitoring butterflies at refaunated and neglected plots. We used correspondence analysis for the presence-absence comparison and canonical correspondence analysis for the current monitoring, and related results of both ordination methods to the life history and climatic traits, and conservation-related attributes, of recorded butterflies.

Results

Following the termination of military use, several poorly mobile species inclining towards oceanic climates were lost. Newly gained are mobile species preferring warmer continental conditions. The refaunated plots hosted higher butterfly species richness and abundances. Larger-bodied butterflies developing on coarse grasses and shrubs inclined towards neglected plots, whereas refaunated plots supported smaller species developing on small forbs.

Conclusion

The changes in species composition following the cessation of military use were attributable to successional change, coupled with changes in species pool operating at larger scales. By blocking succession, large ungulates support butterflies depending on competitively poor plants. Restoring large ungulates populations represents a great hope for conserving specialised insects, provided that settings of the projects, and locally adapted ungulate densities, do not deplete resources for species with often contrasting requirements.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12862-021-01804-x.

A new comprehensive trait database of European and Maghreb butterflies, Papilionoidea

Trait-based analyses explaining the different responses of species and communities to environmental changes are increasing in frequency. European butterflies are an indicator group that responds rapidly to environmental changes with extensive citizen science contributions to documenting changes of abundance and distribution. Species traits have been used to explain long- and short-term responses to climate, land-use and vegetation changes. Studies are often characterised by limited trait sets being used, with risks that the relative roles of different traits are not fully explored. Butterfly trait information is dispersed amongst various sources and descriptions sometimes differ between sources. We have therefore drawn together multiple information sets to provide a comprehensive trait database covering 542 taxa and 25 traits described by 217 variables and sub-states of the butterflies of Europe and Maghreb (northwest Africa) which should serve for improved trait-based ecological, conservation-related, phylogeographic and evolutionary studies of this group of insects. We provide this data in two forms; the basic data and as processed continuous and multinomial data, to enhance its potential usage.

Measurement(s)	resources • Egg Laying • larval environment • pupal environment • geographic location • behavior • size • voltinism • phenology • host plant
Technology Type(s)	digital curation
Factor Type(s)	species
Sample Characteristic - Organism	Papilionoidea
Sample Characteristic - Location	Europe • Northwest Africa

Machine-accessible metadata file describing the reported data: 10.6084/m9.figshare.12998828

How long is 3 km for a butterfly? Ecological constraints and functional traits explain high mitochondrial genetic diversity between Sicily and the Italian Peninsula

Populations inhabiting Mediterranean islands often show contrasting genetic lineages, even on islands that were connected to the mainland during glacial maxima. This pattern is generated by forces acting in historical and contemporary times. Understanding these phenomena requires comparative studies integrating genetic structure, functional traits and dispersal constraints. Using as a model the butterfly species living across the Messina strait (3 km wide) separating Sicily from the Italian Peninsula, we aimed to unravel the mechanisms limiting the dispersal of matrilines and generating genetic differentiation across a narrow sea strait. We analysed the mitochondrial COI gene of 84 butterfly species out of 90 documented in Sicily and compared them with populations from the neighbouring southern Italian Peninsula (1,398 sequences) and from the entire Palaearctic region (8,093 sequences). For each species, we regressed 13 functional traits and 2 ecological constraints to dispersal (winds experienced at the strait and climatic suitability) against genetic differentiation between Sicily and Italian Peninsula to understand the factors limiting dispersal. More than a third of the species showed different haplogroups across the strait and most of them also represented endemic haplogroups for this island. One fifth of Sicilian populations (and 32.3% of endemic lineages) had their closest relatives in distant areas, instead of the neighbouring Italian Peninsula, which suggests high relictuality. Haplotype diversity was significantly explained by the length of the flight period, an intrinsic phenology trait, while genetic differentiation was explained by both intrinsic traits (wingspan and degree of generalism) and contemporary local constraints (winds experienced at the strait and climatic suitability). A relatively narrow sea strait can produce considerable differentiation among butterfly matrilines and this phenomenon showed a largely deterministic fingerprint. Because of unfavourable winds, populations of the less dispersive Sicilian butterflies tended to differentiate into endemic variants or to maintain relict populations. Understanding these phenomena required the integration of DNA sequences, species traits and physical constraints for a large taxon at continental scale. Future studies may reveal if the patterns here shown for mitochondrial DNA are also reflected in the nuclear genome or, alternatively, are the product of limited female dispersal.

Predicting resilience of ecosystem functioning from co-varying species' responses to environmental change

Understanding how environmental change affects ecosystem function delivery is of primary importance for fundamental and applied ecology. Current approaches focus on single environmental driver effects on communities, mediated by individual response traits. Data limitations present constraints in scaling up this approach to predict the impacts of multivariate environmental change on ecosystem functioning. We present a more holistic approach to determine ecosystem function resilience, using long-term monitoring data to analyze the aggregate impact of multiple historic environmental drivers on species' population dynamics. By assessing covariation in population dynamics between pairs of species, we identify which species respond most synchronously to environmental change and allocate species into "response guilds." We then use "production functions" combining trait data to estimate the relative roles of species to ecosystem functions. We quantify the correlation between response guilds and production functions, assessing the resilience of ecosystem functioning to environmental change, with asynchronous dynamics of species in the same functional guild expected to lead to more stable ecosystem functioning. Testing this method using data for butterflies collected over four decades in the United Kingdom, we find three ecosystem functions (resource provisioning, wildflower pollination, and aesthetic cultural value) appear relatively robust, with functionally important species dispersed across response guilds, suggesting more stable ecosystem functioning. Additionally, by relating genetic distances to response guilds we assess the heritability of responses to environmental change. Our results suggest it may be feasible to infer population responses of butterflies to environmental change based on phylogeny-a useful insight for conservation management of rare species with limited population monitoring data. Our approach holds promise for overcoming the impasse in predicting the responses of ecosystem functions to environmental change. Quantifying co-varying species' responses to multivariate environmental change should enable us to significantly advance our predictions of ecosystem function resilience and enable proactive ecosystem management.

Trait-based ecology of terrestrial arthropods

In focusing on how organisms' generalizable functional properties (traits) interact mechanistically with environments across spatial scales and levels of biological organization, trait-based approaches provide a powerful framework for attaining synthesis, generality and prediction. Trait-based research has considerably improved understanding of the assembly, structure and functioning of plant communities. Further advances in ecology may be achieved by exploring the trait-environment relationships of non-sessile, heterotrophic organisms such as terrestrial arthropods, which are geographically ubiquitous, ecologically diverse, and often important functional components of ecosystems. Trait-based studies and trait databases have recently been compiled for groups such as ants, bees, beetles, butterflies, spiders and many others; however, the explicit justification, conceptual framework, and primary-evidence base for the burgeoning field of 'terrestrial arthropod trait-based ecology' have not been well established. Consequently, there is some confusion over the scope and relevance of this field, as well as a tendency for studies to overlook important assumptions of the trait-based approach. Here we aim to provide a broad and accessible overview of the trait-based ecology of terrestrial arthropods. We first define and illustrate foundational concepts in trait-based ecology with respect to terrestrial arthropods, and justify the application of trait-based approaches to the study of their ecology. Next, we review studies in community ecology where trait-based approaches have been used to elucidate how assembly processes for terrestrial arthropod communities are influenced by niche filtering along environmental gradients (e.g. climatic, structural, and land-use gradients) and by abiotic and biotic disturbances (e.g. fire, floods, and biological invasions). We also review studies in ecosystem ecology where trait-based approaches have been used to investigate biodiversity-ecosystem function relationships: how the functional diversity of arthropod communities relates to a host of ecosystem functions and services that they mediate, such as decomposition, pollination and predation. We then suggest how future work can address fundamental assumptions and limitations by investigating trait functionality and the effects of intraspecific variation, assessing the potential for sampling methods to bias the traits and trait values observed, and enhancing the quality and consolidation of trait information in databases. A roadmap to guide observational trait-based studies is also presented. Lastly, we highlight new areas where trait-based studies on terrestrial arthropods are well positioned to advance ecological understanding and application. These include examining the roles of competitive, non-competitive and (multi-)trophic interactions in shaping coexistence, and macro-scaling trait-environment relationships to explain and predict patterns in biodiversity and ecosystem functions across space and time. We hope this review will spur and guide future applications of the trait-based framework to advance ecological insights from the most diverse eukaryotic organisms on Earth.