Epiphytic diatom community structure and richness is determined by macroalgal host and location in the South Shetland Islands (Antarctica)

The second skin of macroalgae: Unveiling the biodiversity of epiphytic microalgae across environmental gradients of the Magellan Subantarctic ecoregion

The Magellan Subantarctic ecoregion (MSE) in the Southern Hemisphere (47°-56°S; 71°-73°W) is a unique natural laboratory subject to persistent and accelerated glacial ice melt, generating a complex system of environmental gradients (e.g., salinity and temperature) that influence the ecological patterns of marine biodiversity. However, the factors influencing marine epiphytic microalgal assemblages are still poorly understood. In this context, we characterized the richness and structure of epiphytic assemblages in different benthic macroalgal hosts (Acrosiphonia arcta, Ectocarpus siliculosus, and Leptosiphonia brodiei) in sites with glaciers and estuarine characteristics (Yendegaia Bay and Fouquet Estuary) and sites without glaciers and oceanic characteristics (Batchelor River and Offing Island) of the MSE, revealing how sites, host, and environmental variables influence variation of epiphytic assemblages. In 36 samples, 67 genera of epiphytes were recorded. The dominant divisions were Bacillariophyta (50 genera), Dinophyta (7 genera) and Cyanophyta (6 genera). We observed significantly high diversity in epiphytic assemblages, with contrasting patterns of variation depending on site and/or host macroalgae. Host specificity was not evident for most epiphytes. The most factor influencing the variation of the epiphythic assemblage was the marked environmental gradient (changes in temperature, salinity, nutrients, among others) between sites with and without glacial influence. Additionally, our research identified potentially toxic and/or harmful epiphytic microalgae belonging to the divisions Dinophyta (dinoflagellates) and Cyanophyta (cyanobacteria). The data on ecological patterns of epiphyte assemblages provides valuable insights into the current state of a poorly understood microscopic biodiversity, shaped by diverse environmental factors at different sites. Under current and future climate change scenarios in the MSE, environmental gradients may become more pronounced, with important positive and/or negative consequences on epiphyte assemblages. In light of these findings, we present a baseline for future research to further develop our understanding and facilitate the monitoring and conservation of epiphytic microalgae in the MSE.

"UDE DIATOMS in the Wild 2024": a new image dataset of freshwater diatoms for training deep learning models

BACKGROUND

Diatoms are microalgae with finely ornamented microscopic silica shells. Their taxonomic identification by light microscopy is routinely used as part of community ecological research as well as ecological status assessment of aquatic ecosystems, and a need for digitalization of these methods has long been recognized. Alongside their high taxonomic and morphological diversity, several other factors make diatoms highly challenging for deep learning-based identification using light microscopy images. These include (i) an unusually high intraclass variability combined with small between-class differences, (ii) a rather different visual appearance of specimens depending on their orientation on the microscope slide, and (iii) the limited availability of diatom experts for accurate taxonomic annotation.

FINDINGS

We present the largest diatom image dataset thus far, aimed at facilitating the application and benchmarking of innovative deep learning methods to the diatom identification problem on realistic research data, "UDE DIATOMS in the Wild 2024." The dataset contains 83,570 images of 611 diatom taxa, 101 of which are represented by at least 100 examples and 144 by at least 50 examples each. We showcase this dataset in 2 innovative analyses that address individual aspects of the above challenges using subclustering to deal with visually heterogeneous classes, out-of-distribution sample detection, and semi-supervised learning.

CONCLUSIONS

The problem of image-based identification of diatoms is both important for environmental research and challenging from the machine learning perspective. By making available the so far largest image dataset, accompanied by innovative analyses, this contribution will facilitate addressing these points by the scientific community.