This dataset seeks to quantify food web flows through contrasting salp-dominated and salp-absent water parcels near the Chatham Rise off western New Zealand where salp blooms are a predictable phenomenon.

Paired 20-liter plankton kreisels or 30-liter pseudo-kreisels were gently filled (using silicon tubing) with mixed layer seawater collected by CTD-Niskin rosette casts at ~22:00 local time. Salps for incubations were collected at ~23:00 local time using the salp net. The net was towed slowly and briefly (5-10 minutes) through the mixed layer. Healthy specimens, i.e., those that showed no physical damage, were then gently transferred (using a large ladle) into a bucket containing filtered surface seawater Specimens were further observed (15-30 minutes) to ensure they actively swam (i.e., that they appeared healthy). They were then transferred into one of the paired kreisels (+Salp treatment), while the second kreisel was used as a control treatment with prey only. We successfully collected and incubated S. thompsoni blastozooids and oozooids ranging in size from 50 – 128 millimeters. We also conducted three incubations with a chain of blastozooids (6-8 millimeter individuals) released by an oozooid inside of one of the plankton kreisels. We found that this was the only way to successfully obtain such small blastozooids in healthy conditions. We also incubated a 112-millimeter Thetys vagina oozooid.

Water was circulated within the kreisels using a peristaltic pump and silicon tubing. Just after salp transfer to the kreisels, initial samples for flow cytometry were taken from each experimental and control kreisel. Additional time points were taken approximately every 2 hours and analyzed in near real time to allow us to monitor salp grazing.Incubations typically lasted ~24 hours.

Samples from the salp incubations were analyzed at sea on a Becton-Dickinson Accuri C6 Plus flow cytometer to estimate the abundance and size of eukaryotic phytoplankton. Samples (~660 µL) were run live within ~1-2 h of collection, discriminating on the Chl a fluorescence signal. We estimated cell diameter from forward light scatter after developing a relationship based on analyses of multiple polystyrene bead sizes (0.99-10 µm diameter). We note that forward scatter is an imperfect proxy for equivalent spherical diameter, and thus the absolute cell sizes determined in our study should be assumed to have associated uncertainty.

For additional details on all methods, see Stukel et al. (in review).

Note: This dataset is also provided in .mat (MATLAB) format under Supplemental Files.