Assessing key influences on the distribution and life-history of Arctic and boreal Calanus: are online databases up to the challenge?

Abstract

Despite the importance of calanoid copepods to healthy ecosystem functioning of the Arctic Ocean and Subarctic Seas, many aspects of their biogeography, particularly in winter months, remain unresolved. At the same time, online databases that digitize species distribution records are growing in popularity as a tool to investigate ecological patterns at macro scales. The value of such databases for Calanus research requires investigation - the long history of Calanus sampling holds promise for such databases, while conditions at high latitudes may impose limits through spatial and temporal biases. We collated records of three Calanus species (C. finmarchicus, C. glacialis, and C. hyperboreus) from the Ocean Biodiversity Information System (OBIS) and the Global Biodiversity Information Facility (GBIF) providing over 230,000 unique records spanning 150 years and over 100 individual datasets. After quality control and cleaning, the latitudinal and vertical distribution of occurrences were explored, as well as the completeness of informative metadata fields. Calanus sampling was found to be temporally and spatially biased towards surfacemost layers (<10m) in spring and summer. Only 3.5% of records had an average collection depth ≥400m, approximately half of these in months important for diapause. Just over 40% of records lacked associated information on sampling protocol while 11% of records lacked life-stage information. OBIS data contained fields for maximum and minimum collection depth and so were subset into discrete “shallow summer” and “deep winter” life cycle phases and matched to sea-ice and temperature conditions. 23% of OBIS records north of 66° latitude were located in regions of seasonal sea-ice presence and occurrences show species-specific thermal optima during the shallow summer period. The collection depth of C. finmarchicus was significantly different to C. hyperboreus during the deep winter. Overall, online databases contain a vast number of Calanus records but sampling biases should be acknowledged when they are used to investigate patterns of biogeography. We advocate efforts to integrate additional data sources within online portals. Particular gaps to be filled by existing or future collections are (i) widening the spatial extent of sampling during spring/summer months, (ii) increasing the frequency of sampling during winter, particularly at depths below 400m, and (iii) improving the quality, quantity and consistency of metadata reporting.