Small herbaria contribute unique biogeographic records to county, locality, and temporal scales

The Dolichens database: the lichen biota of the Dolomites

The Dolichens project provides the first dynamic inventory of the lichens of the Dolomites (Eastern Alps, Italy). Occurrence records were retrieved from published and grey literature, reviewed herbaria, unpublished records collected by the authors, and new sampling campaigns, covering a period from 1820 to 2022. Currently, the dataset contains 56,251 records, referring to 1,719 infrageneric taxa, reported from 1820 to 2022, from hilly to nival belts, and corresponding to about half of the species known for the whole Alpine chain. Amongst them, 98% are georeferenced, although most of them were georeferenced a posteriori. The dataset is available through the Global Biodiversity Information Facility (GBIF; https://www.gbif.org/es/dataset/cea3ee2c-1ff1-4f8e-bb37-a99600cb4134) and through the Dolichens website (https://italic.units.it/dolichens/). We expect that this open floristic inventory will contribute to tracking the lichen diversity of the Dolomites over the past 200 years, and providing the basis for future taxonomic, biogeographical, and ecological studies.

From Dormant Collections to Repositories for the Study of Habitat Changes: The Importance of Herbaria in Modern Life Sciences

In recent decades, the advent of new technologies for massive and automatized digitization, together with the availability of new methods for DNA sequencing, strongly increased the interest and relevance of herbarium collections for the study of plant biodiversity and evolution. These new approaches prompted new projects aimed at the creation of a large dataset of molecular and phenological data. This review discusses new challenges and opportunities for herbaria in the context of the numerous national projects that are currently ongoing, prompting the study of herbarium specimens for the understanding of biodiversity loss and habitat shifts as a consequence of climate changes and habitat destruction due to human activities. With regard to this, the National Biodiversity Future Center (active in Italy since 2022) started a large-scale digitization project of the Herbarium Centrale Italicum in Florence (Italy), which is the most important Italian botanical collection, consisting of more than 4 million samples at present.

Target capture and genome skimming for plant diversity studies

Recent technological advances in long-read high-throughput sequencing and assembly methods have facilitated the generation of annotated chromosome-scale whole-genome sequence data for evolutionary studies; however, generating such data can still be difficult for many plant species. For example, obtaining high-molecular-weight DNA is typically impossible for samples in historical herbarium collections, which often have degraded DNA. The need to fast-freeze newly collected living samples to conserve high-quality DNA can be complicated when plants are only found in remote areas. Therefore, short-read reduced-genome representations, such as target capture and genome skimming, remain important for evolutionary studies. Here, we review the pros and cons of each technique for non-model plant taxa. We provide guidance related to logistics, budget, the genomic resources previously available for the target clade, and the nature of the study. Furthermore, we assess the available bioinformatic analyses, detailing best practices and pitfalls, and suggest pathways to combine newly generated data with legacy data. Finally, we explore the possible downstream analyses allowed by the type of data generated using each technique. We provide a practical guide to help researchers make the best-informed choice regarding reduced genome representation for evolutionary studies of non-model plants in cases where whole-genome sequencing remains impractical.

How does spatial extent and environmental limits affect the accuracy of species richness estimates from ecological niche models? A case study with North American Pinaceae and Cactaceae

Measuring species richness at varying spatial extents can be challenging, especially at large extents where exhaustive species surveys are difficult or impossible. Our work aimed at determining the reliability of species richness estimates from stacked ecological niche models at different spatial extents for taxonomic groups with vastly different environmental dependencies and interactions. To accomplish this, we generated ecological niche models for the species of Cactaceae and Pinaceae that occur within 180 published floras from North America north of Mexico. We overlaid or stacked the resulting species' potential distribution estimates over the bounding boxes representing each of the 180 floras to generate predictions of species richness. In general, our stacked models of Cactaceae and Pinaceae were poor predictors of species richness. The relationships between observed and predicted values improved noticeably with the size of spatial extents. However, the stacked models tended to overpredict the richness of Cactaceae and over- and underpredict the richness of Pinaceae. Cactaceae stacked models showed higher sensitivity and lower specificity than those for Pinaceae. We conclude that stacked ecological niche models may be somewhat poor predictors of species richness at smaller spatial extents and should be used with caution for this purpose. Perhaps more importantly, abilities to compensate for their limitations or apply corrections to their reliability may vary with taxonomic groups.

The Italian lichens dataset from the TSB herbarium (University of Trieste)

Background

The "Herbarium Universitatis Tergestinae" (TSB), with a total of ca. 50,000 specimens, includes the largest modern collection of lichens in Italy, with 25,796 samples collected from all over the country since 1984, representing 74% of all taxa known to occur in Italy. Almost all specimens have been georeferenced "a posteriori". The dataset is available through GBIF, as well as in ITALIC, the Information System of Italian Lichens.

New information

The TSB Herbarium hosts the largest modern lichen collection in Italy, with a total of ca. 50,000 specimens. This dataset contains all of the 25,796 specimens collected within the administrative borders of Italy. Amongst them, 98% are georeferenced and 87% have the date of collection. The dataset includes several type specimens (isotypes and holotypes) and exsiccata.

Conservation ethics in the time of the pandemic: Does increasing remote access advance social justice?

The COVID-19 pandemic is stimulating improvements in remote access and use of technology in conservation-related programs and research. In many cases, organizations have intended for remote engagement to benefit groups that have been marginalized in the sciences. But are they? It is important to consider how remote access affects social justice in conservation biology-i.e., the principle that all people should be equally respected and valued in conservation organizations, programs, projects, and practices. To support such consideration, we describe a typology of justice-oriented principles that can be used to examine social justice in a range of conservation activities. We apply this typology to three conservation areas: (1) remote access to US national park educational programs and data; (2) digitization of natural history specimens and their use in conservation research; and (3) remote engagement in conservation-oriented citizen science. We then address the questions: Which justice-oriented principles are salient in which conservation contexts or activities? How can those principles be best realized in those contexts or activities? In each of the three areas we examined, remote access increased participation, but access and benefits were not equally distributed and unanticipated consequences have not been adequately addressed. We identify steps that can and are being taken to advance social justice in conservation, such as assessing programs to determine if they are achieving their stated social justice-oriented aims and revising initiatives as needed. The framework that we present could be used to assess the social justice dimensions of many conservation programs, institutions, practices, and policies.

Harnessing Large-Scale Herbarium Image Datasets Through Representation Learning

The mobilization of large-scale datasets of specimen images and metadata through herbarium digitization provide a rich environment for the application and development of machine learning techniques. However, limited access to computational resources and uneven progress in digitization, especially for small herbaria, still present barriers to the wide adoption of these new technologies. Using deep learning to extract representations of herbarium specimens useful for a wide variety of applications, so-called "representation learning," could help remove these barriers. Despite its recent popularity for camera trap and natural world images, representation learning is not yet as popular for herbarium specimen images. We investigated the potential of representation learning with specimen images by building three neural networks using a publicly available dataset of over 2 million specimen images spanning multiple continents and institutions. We compared the extracted representations and tested their performance in application tasks relevant to research carried out with herbarium specimens. We found a triplet network, a type of neural network that learns distances between images, produced representations that transferred the best across all applications investigated. Our results demonstrate that it is possible to learn representations of specimen images useful in different applications, and we identify some further steps that we believe are necessary for representation learning to harness the rich information held in the worlds' herbaria.