Many leading journals in ecology and evolution now mandate open data upon publication. Yet, there is very little oversight to ensure the completeness and reusability of archived datasets, and we currently have a poor understanding of the factors associated with high-quality data sharing. We assessed 362 open datasets linked to first- or senior-authored papers published by 100 principal investigators (PIs) in the fields of ecology and evolution over a period of 7 years to identify predictors of data completeness and reusability (data archiving quality). Datasets scored low on these metrics: 56.4% were complete and 45.9% were reusable. Data reusability, but not completeness, was slightly higher for more recently archived datasets and PIs with less seniority. Journal open data policy, PI gender and PI corresponding author status were unrelated to data archiving quality. However, PI identity explained a large proportion of the variance in data completeness (27.8%) and reusability (22.0%), indicating consistent inter-individual differences in data sharing practices by PIs across time and contexts. Several PIs consistently shared data of either high or low archiving quality, but most PIs were inconsistent in how well they shared. One explanation for the high intra-individual variation we observed is that PIs often conduct research through students and postdoctoral researchers, who may be responsible for the data collection, curation and archiving. Levels of data literacy vary among trainees and PIs may not regularly perform quality control over archived files. Our findings suggest that research data management training and culture within a PI's group are likely to be more important determinants of data archiving quality than other factors such as a journal's open data policy. Greater incentives and training for individual researchers at all career stages could improve data sharing practices and enhance data transparency and reusability.
We assessed the quality of 362 open datasets shared by 100 principal investigators (PIs) in ecology and evolution to identify predictors of data quality. Datasets generally scored low on completeness and reusability, but these metrics were slightly higher for more recently archived datasets and PIs with less seniority. Journal data sharing policies had no effect on data quality, whereas PI identity explained the largest proportion of the variance in both data completeness (27.8%) and reusability (22.0%), suggesting that a PI’s training and lab culture are key determinants of data quality. Thus, greater incentives and training for individual researchers could help improve data sharing practices.
Spatiotemporal overlap between fish larvae and their planktonic prey is an important source of recruitment variability. Over the past decade, one species of redfish, Sebastes mentella, from the Gulf of St. Lawrence (GSL) produced multiple strong cohorts following decades of low recruitment, which has generated strong interest in identifying potential drivers of larval survival. The present study provides the first detailed, multi-year assessment of larval redfish (Sebastes spp.) trophodynamics. Interannual variability in larval redfish diet composition and prey selectivity was assessed using high-resolution prey identification of larval gut contents and in situ prey fields. Eggs from the calanoid copepod Calanus finmarchicus represented the most frequently consumed prey in 3 of the 4 collection years, and contributed the largest proportion of carbon ingested by redfish larvae in all years. The high consumption of C. finmarchicus eggs by larvae, combined with evidence of positive selection for this taxon in some years, supports the hypothesis of a strong trophic link between larval redfish and a key calanoid copepod in the GSL ecosystem. Our results indicate that future efforts investigating GSL redfish recruitment processes should consider environment-driven variability in the reproductive phenology and abundance of C. finmarchicus.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.