1. CTD and rosette CLOSING REMARKS Think about huge range of options, e.g. for measuring temperature:

2. microcat

3. XBT: expendable profiling temperature sensor, profile depth normally 800m

4. Interdisciplinary sensors: For autonomous moored applications, sensors for many variables are available now 14 C Primary Production Measurements(C. Taylor) Optical (Dickey) and O 2 sensors (Wanninkhof) CO 2 sensor (M. DeGrandpre)

5. Where should we go Minituarize sensors: Biogeochemical sensors that are small, low-power, “dry” (optical, acoustic, chips) meteorological sensors that can be submerged Micro-humidity sensor (JPL) Optical O 2 sensor MEMS chip

6. Where should we go 3 rd generation autonomous vehicles (symbiosis of glider, float, AUV) : ability to travel long distances to mission area and back to shore base (propelled or glider mode) choice of going into float mode, glider mode (sections) or mooring mode long endurance (thermal or solar power) option to receive sound signals for tracking (Rafos), e.g. under ice, and tomography flexible choice of sensors ability to collect ml water samples and bring home for analysis Autonomous surface craft are also receiving more attention now:

7. “Liquid Robotics” waver glider http://liquidr.com/

9. Where should we go Multidisciplinary moorings for - high-frequency observations - strong current regimes - process studies - heavy/large sensors - reference sites - sound sources long-life, advanced telemetry small, cheap, expendable ? deployable from VOS or autonomous craft ? docking of gliders for calibration and to return samples self-calibrating sensors

10. Sampling characteristics of the platforms Research vessels - can take samples - very sparse sampling ($25000/day) - handle/deploy heavy equipment - expensive (too much for operational - reach remote areas (use like VOS) obs, but needed for servicing) Platform strengths weaknesses VOS - high resolution along repeat tracks - tracks not always where wanted (free) - for surface reading many variables - tracks may change, they don‘t stop - no subsurface except T (800m) Surface drifters - global coverage - sparse spatial sampling ($2000 ?) - rapid sampling in time - only surface obs - low-cost, robust technology - limited variables (T, air p, S) Floats - global coverage - coarse x,y,t resolution ($15000+5000) - vertical profiling to mid-depth - limited weight/power for sensors - “cheap“ so large numbers feasible - avoid or quickly leave certain regions Moorings - high time resolution, surface to bottom - no x,y resolution ($250000) - many variables possible - expensive, incl. ships needed - can monitor adverse/difficult locations - large technical effort/few groups - re-calibrations, so can be reference Gliders - good sampling along tracks - very slow (20-25cm/s) ($70000) - free choice of track, can be steered - limited depth range and variables - small sensor suite feasible Integrals - integrate over long distances - expensive - good time resolution - limited variables and places possible Coastal radars - good x,y,t resolution - limited coverage - land based - only surface, only currents and waves

11. moorings gliders satellites floats VOS radars mooring networks