5. Missions execution

We recommend as many nearby CTD casts with water samples for high accuracy lab based nitrate or nitrate+nitrite measurements as possible. Measurements of temperature, salinity and turbidity are required, especially in areas with high gradients. Dissolved and particulate organic matter measurements can be useful for validation of UV spectra. If there is no cross-comparison and no control of the environmental interferences variation, the uncertainty of the measurement must be increased.

5.1. UV Spectrometers

For nitrate measuring UV spectrometers glider mission execution is fairly straightforward. As there are no sensor delays or similar to consider, the main point in mission planning and execution is energy consumption and possibly minimization of bio-fouling. Since these instruments use a significant amount of power the need for vertical resolution has to be balanced against the length of the deployment.

In difference to e.g. Oxygen optodes the measurement of only down- or upcasts is a possibility as there is no reason for them to differ (optodes have significant lags). For Deep-SUNAs with their relatively low noise levels one measurement every few seconds appears sufficient to resolve the larger scale nitrate concentration variations.

On Slocum, seaglider and seaexplorer gliders it is also possible to run different sample rates for different depth ranges. At GEOMAR reduced sampling rates have occasionally been used at depths below 500 m where nitrate concentrations were relatively constant.

When programming the Deep SUNA there are several options for how the data is stored internally. Storing of reduced spectra has also shown to be problematic and storage space will not likely be a limiting factor for normal operations.

We strongly recommend daily storage files and internal storage of full spectra. Data transmitted to Slocum gliders via the serial connection must however be ‘reduced spectra’ as the glider processor cannot handle more data.

5.2. Lab-on-Chip instruments

The relatively slow sampling frequency of the LoC analyser means that following ‘standard’ dive flight paths results in a small number of samples in shallow waters (e.g. 4-5 in 150 m water column; [Vincent et al., 2018]). In such circumstances a ‘loiter’ profile is preferred, allowing a greater number of NO₃ measurements (e.g. 10 in 150 m water column; [Vincent et al., 2018]). Using a glider with an unpumped CTD package operating in loiter mode can result in slow and variable flow through the conductivity cell and result in suboptimal conductivity measurements when passing through temperature gradients. Therefore, when using unpumped CT the loiter dive must be paired with a standard profile and the CT values extrapolated to the loiter. Pumped CTD packages should ideally be used to overcome this issue.