Assessing the Quality of iNaturalist Spider (Araneae): Observations and the iImportance of using metadata and Image quality control when using crowd-sourced databases for research

Crowd-sourced biodiversity databases like iNaturalist have grown in popularity in the past 10 years. Publicly sourced biodiversity databases provide researchers with datasets and images for educational and research purposes. The data on these sites has proven useful in biodiversity, conservation, and ecological research. One concern with using publicly sourced databases, however, is the quality of both the metadata and images obtained from the sites. I evaluated metadata and images from iNaturalist, a global crowd-sourced biodiversity website, for observations of spiders (order Araneae) to see if the current collection of observations (both images and metadata) on iNaturalist can provide a reliable source for research data and images. The observations for the order Araneae were evaluated on the diversity of observations, quality of associated images, as well as for the level and accuracy of the taxonomic identifications, accuracy of geo-tagging, and number of duplications. I used BRISQUE image quality assessment algorithm to evaluation the photographs.  I found that iNaturalist currently contains at least one observation for 122 of the 129 families of Araneae, and 115 families had three or more unique observations with relatively consistent accuracy for the observation metadata and image quality between the families. Additionally, the methods described present a possible means of assessing and quantifying crowd-sourced data used for research purposes.