Sources of variability in geographical and seasonal differences in particulate fluxes from short-term sediment trap deployments, east of New Zealand

Free-floating sediment traps were used to measure total mass and particulat e phosphorus fluxes from the upper ocean (>100-550 m) at two stations in ea ch of three water types (subantarctic, Subtropical Convergence and subtropi cal), east of New Zealand, in winter and spring 1993. Despite marked change s in ecosystem community structure and function between water types and occ asionally between stations, there were no significant differences in export flux either across water masses, between stations or with increasing water depth. One exception was in the Subtropical Convergence where significant increases in particulate flux with depth at one station in spring are infer red to be caused by tidal current resuspension of bottom sediments on the c rest of the Chatham Rise. High degrees of variation between mass fluxes cal culated from individual cylinders observed in winter (coefficients of varia tion ranging from 10 to 123%) may have been due to inter-trap hydrodynamic interactions or sample processing errors. In spring, overall coefficients o f variation of mass flux were smaller than in winter (6-87%), but variation s due solely to sub-sampling procedures were significant (<1-83%). The rang es of variability for the New Zealand traps are similar to other published sediment trap studies, suggesting that the use of single, free-floating sed iment trap arrays to characterise fluxes for specific oceanic provinces is unlikely to be valid statistically. Based on the results from east of New Z ealand, more than two arrays would be required to improve the pow er of the chosen statistical test in order to determine significant differences in p articulate flux between physically and biologically distinctive oceanic wat er masses. (C) 1998 Elsevier Science Ltd. All rights reserved.