The IDC student paper awards 2025 have been awarded to Friederike Barkmann, Valentin Johannes Heimer, Benedikt Hiebl, Carlotta Musso, Fabrizzio Protti Sánchez, Maria Chiara Vulcano, Sebastiano Zanini and Anna de Vries.

Congratulations!

Below you find a description of the awarded papers.

Mapping Butterfly Species Richness and Abundance in Mountain Grasslands —Spatial Application of a Biodiversity Indicator
Friederike Barkmann, Erich Tasser, Ulrike Tappeiner, Peter Huemer, Benjamin Schattanek-Wiesmair, Kurt Lechner, Alois Ortner and Johannes Rüdisser

Butterflies are not only charismatic insects but also important biodiversity indicators that react sensitively to changes in environmental conditions. Information on their species richness and abundance in grassland habitats can be used to inform policy makers, raise awareness in the public and plan and evaluate nature protection measures. We used data from over 175 butterfly survey sites from the Viel-Falter monitoring and environmental variables to predict butterfly species richness and abundance for all grassland habitats in Vorarlberg and Tyrol. The resulting maps of predicted species richness and abundance provide assessments beyond the limited number of survey sites at a high spatial resolution which is especially relevant for local
and regional management actions. We also assessed how factors at local and landscape level affect butterfly populations. In the mountainous study region elevation was an important predictor with highest species richness and abundance at mid-elevations. In addition, there was a positive effect of south facing slopes, medium levels of productivity and forest edges in the surrounding landscape. The study shows how the integration of field surveys with high resolution remote sensing data can improve our understanding of spatial biodiversity patterns and the drivers that influence them.

Machine learning training data: over 500,000 images of butterflies and moths (Lepidoptera) with species labels
Friederike Barkmann, Andreas Lindner, Ronald Würflinger, Helmut Höttinger, Johannes Rüdisser

Apps to share sightings of species by taking and uploading pictures have considerably gained popularity in the last years resulting in large datasets of species records. The collected information on the occurrences of different species can contribute to a better understanding of species distributions and ecosystem dynamics but the processing of the datasets can be time-consuming and expensive. Deep learning models (artificial neural networks) can support quality control by automatically identifying species on the images but they require large training datasets for reliable classification. We published a dataset with over 500,000 images of butterflies and moths that were taken by users of the App “Schmetterlinge Österreichs” of the foundation “Blühendes Österreich”. The correct identification of the 185 species in the dataset was validated by an expert. A model that was trained on the dataset could correctly predict nearly 98 % of the images. The model did not work equally well for all species though, with lower performance for some species that were represented by fewer images. The published dataset is a valuable resource to train and test other models on a species identification task and compare different methods for automated species identification. In addition, it provides the necessary foundation to establish automated species identification models for most Austrian butterfly species which can be used for scientific and educational purposes.

Disentangling the cryptic evolution of European woodrushes

Heimer V., Carnicero P., Carrizo García C., Hilpold, A., Dolenc Koce J., Leal J. L., Li M., Varotto C., Schönswetter P. & Frajman B. (2025). Hybridization and Polyploidy Shaped the Evolutionary History of a Complex of Cryptic Species in European Woodrushes (Luzula sect. Luzula). Systematic Biology, syaf065. (DOI)

Polyploidization – the duplication of entire genomes – is a major driver of plant evolution, but it also complicates the reconstruction of evolutionary relationships, especially in groups shaped by repeated hybridization. European woodrushes (Luzula sect. Luzula) exemplify this challenge: they form a taxonomically complex group with widespread polyploidy, extensive hybridization, and strikingly uniform morphology. As a result, their evolutionary history has remained poorly understood, even in well-studied regions such as the Alps.

Focusing on the Eastern Alps, a center of diversity for this group, we investigated the evolutionary relationships among nine Luzula species using an exceptionally extensive sampling of more than 1000 individuals. We combined genome-wide sequencing data (ddRADseq) with plastid DNA sequences, relative genome size measurements, and chromosome counts in an integrative analytical framework. This approach allowed us to resolve complex, reticulate relationships that could not be disentangled using any single method alone.

Our analyses reveal weak genetic differentiation and signatures of hybridization among several diploid lineages. Building on this foundation, we reconstructed the origins of polyploid species and identified key hybridization events associated with chromosome duplication (allopolyploidy). Notably, we provide the first empirical evidence of natural hybridization between a diploid and an allotetraploid species resulting in a stable tetraploid species.

Our results demonstrate that repeated whole-genome duplications and pervasive hybridization – both within and across ploidy levels – have been central to the diversification of this cryptic species complex and showcase the power of integrative approaches for resolving complex evolutionary histories.

Schematic representation of the most likely evolutionary scenario (a) and a plausible alternative (b) for the polyploid complex of European Luzula sect. Luzula. Ploidy levels are indicated on the left and separated by dotted horizontal lines. Karyotypes are depicted schematically with colors representing ancestral diploid subgenomes. Black arrows show proposed hybridization events, and dashed arrows represent evolutionary divergence without hybridization. Analyses supporting or objecting the involvement of each species in a given hybridization event are indicated as colored squares.

Advancing forest mapping: Pretraining strategies and deep-ensemble based uncertainty for predicting evergreen broad-leaved cover from Sentinel-2 time series

Benedikt Hiebl, Nicola Alessi, Giacomo Calvia, Alessandro Bricca, Gianmaria Bonari, Giulio Zangari, Wouter Dorigo, Stefan Zerbe, Martin Rutzinger

Mediterranean and temperate forests are currently undergoing changes due to climate change and land-use dynamics. In particular, evergreen broad-leaved tree species are expanding into regions previously dominated by deciduous forests. Yet, mapping these shifts at a functional level over large areas remains challenging due to limited availability of accurate field observations, which constrains data-driven Machine Learning approaches.

In this study, we investigated deep learning strategies to map the fractional cover of evergreen broad-leaved trees using annual dense Sentinel-2 satellite time series. We show that pretraining a Convolutional Neural Network on large, contextually similar datasets—using existing forest vegetation databases—substantially improves prediction accuracy when only limited field data for a specific mapping target are available.

Beyond improved accuracy, we explicitly quantify prediction uncertainty using deep ensembles. This allows us to identify regions where the model is less confident, such as areas with forest types not well represented in the training data. The resulting maps therefore not only provide estimates of evergreen broad-leaved forest cover, but also indicate where predictions should be interpreted with caution.

Overall, the study demonstrates that combining transfer learning with uncertainty-aware deep learning enables more reliable and transparent forest mapping from satellite time series under data scarce conditions, supporting biodiversity assessments and climate adaptation strategies.

Global variability in hydraulic traits and water use strategies of mountain shrubs and dwarf shrubs

Musso C., Ganthaler A., Mayr S.

Shrubs are woody plants that grow more in width than in height and can survive even in harsh high‑mountain environments. Although they are an important growth form in these areas, their hydraulic properties (water uptake, transport, and release) remain poorly understood. In this study, we reviewed the water‑use strategies of mountain shrubs across different climates and regions. Our analysis shows that shrubs rely on diverse hydraulic strategies to maintain a favourable water status. However, important knowledge gaps remain, particularly regarding root traits, embolism recovery, and water storage. This review provides new insights into how shrubs adapt to the environmental stresses of mountain regions.

Soil respiration responses to prolonged warming vary seasonally in a subarctic grassland

Fabrizzio Protti-Sánchez, Linsey M. Avila, Kathiravan Meeran, Johannes Ingrisch, Páll Sigurðsson, Klaus S. Larsen, Bjarni D. Sigurdsson, Ivan Janssens, Michael Bahn

Northern ecosystems are warming faster than most other regions, yet how this affects the large amounts of carbon stored in their soils remains uncertain. A key process by which soil carbon is released to the atmosphere is soil respiration – the CO2 produced by soil microbes and roots. This process is highly temperature-sensitive and is expected to increase under climate warming. However, whether this effect persists over long timescales and across seasons remains unclear.
In this study, we used a natural geothermal warming gradient in a subarctic grassland in Iceland, where soil temperatures range from ambient to more than 10 °C above ambient, to investigate how more than a decade of warming influences soil respiration throughout the year. We found that warming significantly increased soil carbon release during the cold season, while having no detectable effect in summer, leading to higher total annual soil carbon emissions. As soils became warmer, they also became less sensitive to further temperature increases, and seasonal contrasts in soil respiration gradually weakened.
Overall, our results highlight the importance of considering seasonal variations when predicting warming impacts on soil carbon dynamics. By providing new insights into how long-term warming influences carbon release in northern ecosystems, this study improves our understanding of their potential to amplify global warming as temperatures continue to rise.

How well do we understand the role of periphyton as a source of dietary energy and nutrients in mountain stream food webs?

Maria Chiara Vulcano, Martin J. Kainz, Georg H. Niedrist, Leopold Füreder

Periphyton is a key basal resource in stream food webs, supplying dietary energy and essential nutrients to aquatic consumers. These nutrients include omega-3 and omega-6 polyunsaturated fatty acids (PUFA), which are essential for the growth, reproduction, and development of aquatic consumers.

While other aspects of mountain stream ecology and hydrology are better studied, the nutritional features of periphyton (both its quantity and nutritional quality) have received little attention. Our literature review revealed that only a few studies have examined the patterns of periphyton biomass in alpine and subalpine streams, and that only six recent studies have assessed its lipid and PUFA content.

Assessing both the quantity and quality of periphyton is crucial because, for example, large amounts of periphyton lacking essential nutrients cannot be efficiently used by consumers, whereas smaller amounts of nutrient-rich periphyton may provide a greater nutritional support. Quantity alone is therefore insufficient to understand trophic dynamics. 

Periphyton quantity and quality can vary seasonally and spatially across alpine and subalpine catchments. For example, available data show that regional climate and environmental conditions drive seasonal patterns of periphyton biomass in high-elevation catchments, and suggests that streams with contrasting nutrient and disturbance regimes can exhibit markedly different biomass levels (see figure).

A deeper understanding of patterns in periphyton quantity and quality is important to explain current trophic dynamics in mountain streams, and for predicting their future development under ongoing climate change.

Periphyton Chlorophyll-a content in alpine glacial and krenal streams of the European Alps. (A) Variations of periphyton Chl-a (used as a measure of periphyton biomass) in glacial and krenal streams over the course of a year. (B) Monthly Chl-a content of periphyton in glacial and krenal streams. Data have been extracted from Uehlinger et al. (2010) using WebPlotDigitizer. We included data from streams surveyed for at least 1 year, with no more than four monthly measurements missing and gaps no longer than two consecutive months.

Disentangling biodiversity and temperature effects on bees and pollination services in mountain agroecosystems

Sebastiano Zanini, Matteo Dainese, Lisa Obwegs, Elia Guariento, Timo Kopf, Matteo Anderle, Georg Leitinger, Ulrike Tappeiner

In agricultural landscapes, wild bees are essential partners for both natural ecosystems and food production. Yet their populations are under pressure from intensive farming practices and climate change. This study explored how local biodiversity (used as an indicator of land-use intensity) and temperature influence wild bees and the pollination they provide in mountain farmland systems in South Tyrol, Italy.

To do this, we selected study sites along two independent gradients and examined how local biodiversity (measured as the diversity of multiple animal and plant groups; also called multidiversity for this reason) and temperature shape wild bees and pollination services. We assessed wild bee diversity using white, blue, and yellow pan traps, recorded honeybee and wild bee flower visits on potted strawberry and radish plants, and measured fruit and seed production. These variables were collected across apple orchards, vineyards, hay meadows, orchard meadows, and pastures.

We found that higher local biodiversity strongly supports wild bee abundance and species richness. In contrast, temperature alone had relatively limited direct effects on wild bee diversity. Wild bee visitation was also higher in biodiverse landscapes, while honeybees tended to dominate flower visits in simplified, low-diversity areas.

Experiments with strawberries and radishes revealed that insect pollination can help buffer the negative effects of warmer conditions on crop production. When pollinators could not pollinate flowers through nets around the plants, higher temperatures led to lower strawberry fruit weight. Instead, pollinated plants were less affected by warming.

Overall, the study highlights the close links between biodiversity, land use, and climate (simulated through elevation along a temperature gradient). It suggests that conserving and restoring biodiversity in agricultural landscapes can strengthen resilience to climate change with an “insurance effect”. Protecting a rich variety of species and habitats not only supports bee diversity, but also help maintain essential ecosystem functions such as pollination, which underpin both wild plant reproduction and food production.

The left side of the figure shows four panels illustrating the relationships between wild bee species richness (SR wild bees) and abundance along two independent environmental gradients: the multidiversity index (left panels) and mean annual temperature (MAT; right panels). The coloured plots on the right side of the figure display how wild bee and honeybee visitation rates vary with the multidiversity index and the number of open radish flowers. Colour bars next to each plot represent visitation intensity. All plots are based on estimates from linear mixed-effects models (LMMs).

DOI

Wetlands -the overlooked COS flux-

Anna de Vries, Georg Wohlfahrt, Kukka‐Maaria Kohonen, Camille Abadie, Marine Remaud, Jürgen Kesselmeier, Asta Laasonen, Mary Whelan, Ivan Mammarella and Timo Vesala

Carbonyl sulfide (COS) is a gas used to study the carbon cycle. This because plants absorb COS in a similar way as they absorb CO2. However, to use COS effectively, all sources and sinks must be understood and nowadays there still is an unexplained missing sink in high northern latitudes. Boreal COS budget calculations typically account for the contribution by forests, but ignore any uptake that wetland ecosystems, widespread in Northern latitudes, may contribute. The first direct measurements of the ecosystem-atmosphere COS exchange of a boreal wetland, presented in this article, demonstrate their likely importance for Northern latitude COS budgets. The investigated wetland (Siikaneva, Finland) took up on average 11 pmol m−2s−1 COS, which was c. 72 % of the nearby boreal forest COS uptake. During nighttime, the COS uptake rates were similar at both sites. Upscaling our measurements to the boreal region using the ORCHIDEE model revealed a Northern wetland sink of c. 13 Gg S/y, changing the simulated budget from a small source to a COS sink impacting Northern latitudes carbon uptake estimates based on COS. With these measurements we show the importance of including boreal wetland COS fluxes in budget calculation.

Highlights

• This study presents the first direct measurements of COS exchange in a boreal wetland (Siikaneva, Finland).
• The wetland showed an average COS uptake of 11 pmol m⁻²s⁻¹, about 72% of the uptake by nearby boreal forests.
• At night, both the wetland and forest showed similar COS uptake.
• When upscaled to the entire boreal region using the ORCHIDEE model, wetlands were estimated to be a COS sink of ~13 Gg S/year. This inclusion shifts the COS budget from a small source to a net sink, impacting carbon uptake estimates

  

de Vries, Anna, et al. "The contribution of boreal wetlands to the Northern Hemisphere carbonyl sulfide sink." Geophysical Research Letters 52.12 (2025): e2024GL112858.