Continuous, long-duration and high resolution hydrographic sampling from gliders have enabled the evaluation of variability in density structure of submesoscale features in a New Zealand shelf sea. Glider sampling allows for: 1) minimal disturbance of upper stratification and 2) horizontal and temporal spacing between profiles to be typically less than one kilometer and 30 minutes depending on the profile depth. Seven glider surveys were completed from 2015 to 2018. The average glider track spanned 132 40^o75S, 174^o49E to 39^o91S, 171^o90E. In each survey, the glider transverses from east to west and back to its deployment location. For surveys 3, 4, 6, 9 and 11, the glider was deployed closer to the Cook Strait Narrows. As the glider would spend multiple days trying to overcome strong currents due to strong tidal fluctuations near the Narrows, deployments for surveys 12 and 15 were from Tasman Bay to maximise observations across the Greater Cook Strait shelf sea.

This study used Teledyne Webb Research Slocum G2 gliders equipped with Seabird CTD sensor, Aanderaa Oxygen Optode and Wet Labs Environmental Characterization Optics (ECO) puck, that measured chlorophyll-a fluorescence, backscatter (at 470, 532, 660 and 700 nm) and chromophoric dissolved organic matter (CDOM). Temperature, conductivity, and pressure data were sampled at 0.5 Hz, and subsequently processed to remove spikes. The accuracy within calibration range of temperature and conductivity were +/-0.002^oC and +/-0.0003 S m^-1, respectively. Glider data processing was completed using the SOCIB glider toolbox (https://github.com/socib/glider_toolbox; Troupin et al. (2015)). Glider data processing includes salinity lag correction for the thermal lag error for the un-pumped CTD unit. Data were averaged in vertical bins of 1 m.