Evaluating shortfalls and spatial accuracy of biodiversity documentation in the Atlantic Forest, the most diverse and threatened Brazilian phytogeographic domain

Abstract

Digital accessible knowledge of biodiversity data is an increasingly important source of information in studies of biogeography and conservation. These databases can also reveal temporal, spatial and taxonomical gaps in biodiversity documentation, even in areas that have been intensively studied and from where accurate species lists are available. Therefore, revealing these gaps may help allocating collecting efforts, conservation priorities and strategies for improving database curation. Here, we evaluate potential shortfalls for flowering plants in a tropical hotspot, the Brazilian Atlantic Forest, by cross‐referencing two online repositories of biodiversity data (the Global Biodiversity Information Facility – GBIF – and the Brazilian Flora 2020 floristic database – BFG). We aimed to evaluate the congruence between those repositories, highlighting tendencies in current documentation for this area. We found that from the 7220 reported flowering plant species endemics to the Atlantic Forest, 1573 (22%) have no valid spatial data in GBIF, and 75% of all of the 605,951 records do not present valid spatial information. Most of the missing information is related to species known only from few and old collections with absent or inaccurately georeferenced data. This lack of information may cause a large impact in spatial studies, especially for rare and threatened species. Nevertheless, our analysis also shows that spatial information for the filtered data is highly congruent between GBIF and BFG data, indicating relatively high availability of quality data in large repositories after standard and automatized cleaning procedures. Still, good practices to decrease the impact of losing data are recommended, including more investment in field collections, targeting poorly known species and returning cleaned spatial datasets to online repositories after taxonomic revisions.

Biodiversity at the global scale: the synthesis continues

“…Not all assessments have necessarily been positive. For instance, misidentifications are pervasive in herbarium specimens, with recent studies finding misidentification rates of >50% (although this impressive number includes correct determinations under outdated taxonomies; Goodwin et al, 2015;Colli-Silva et al, 2020), consistent with our own monographic experience and with estimates in other taxa (Freitas et al, 2020). While we suggest optimism overall, both positive and negative results are probably scale-and hypothesis-dependent.…”

Plant and fungal collections: Current status, future perspectives

“…(2016) note that locally available funding and participation in data sharing networks facilitate mobilization of data and Colli‐Silva et al. (2020) provide a Brazilian example of this. Citizen science platforms such as iNaturalist (Heberling & Isaac, 2018) and the growth of observation data in GBIF demonstrate the potential for increasing the number of recorded occurrences and enable a broader audience to participate in annotating the specimen regarding its identity or other attributes of the organism, and enhancing or correcting the data provided with the original collection.…”

Vellozo's Florae Fluminensis: A new assessment of the São Paulo part of his collecting itinerary, its vegetation, and species list

“…Also, it is worth to mention another relevant element in this regard: the lure of the bait in the name of Florae Fluminensis itself, leading to a misconception that all of the FF names are from the Atlantic Forest domain in Rio de Janeiro State. It is therefore not surprising to find in GBIF a number of Vellozo's names with “missing data”, calling for a re‐evaluation of such gaps in spatial accuracy of documentation of the biodiversity for the Atlantic Forest (Colli‐Silva & al., 2020).…”