User`s manual

Manual revision 016 Appendix V: Real-Time Data Correction Algorithms SBE 49

Appendix V:

Real-Time Data Correction Algorithms

Aligning, filtering, and correcting for conductivity cell thermal mass in post-

processing allows you to make adjustments to the measured data before

calculating derived parameters. SBE Data Processing, part of our Seasoft suite,

can perform this processing on recorded data, if a complete C, T, and P time

series is recorded. However, the FastCAT, which samples and transmits data at

16 Hz, is frequently integrated by the customer with an AUV or ROV. The

AUV/ROV may record data on command, with uneven time intervals between

each recorded measurement. The lack of a true time series from the FastCAT for

these applications prevents aligning, filtering, and correcting for cell thermal

mass in post-processing.

Therefore, the FastCAT can be programmed to process data in real-time if

sampling autonomously. This provides T, C, and P output that is already

aligned, filtered, and corrected for cell thermal mass, and also provides

calculated and output salinity and sound velocity (if applicable) based on these

corrected values.

Cell thermal mass corrections should not be applied to freshwater data. It

can give bad results, due to the way the derivative dc/dT is calculated in

regions where conductivity changes are small. Currently, the SBE 49

firmware does not permit you to apply just aligning and filtering, without also

applying cell thermal mass corrections.

Aligning Data

The FastCAT aligns temperature data in time, relative to conductivity and

pressure. This ensures that calculations of salinity, sound velocity, and other

derived variables are made using measurements from the same parcel of water.

There are three principal causes of misalignment of CTD measurements:

• physical misalignment of the sensors in depth

• inherent time delay (time constants) of the sensor responses

• water transit time delay in the pumped plumbing line - the time it takes

the parcel of water to go through the plumbing to each sensor (or, for free-

flushing sensors, the corresponding flushing delay, which depends on

profiling speed)

In the FastCAT, temperature data is usually misaligned with respect to

conductivity and pressure. Advancing temperature relative to conductivity and

pressure can compensate. The recommended temperature advance value for a

typical FastCAT is 0.0625 seconds, but the user can select a value ranging

from 0 to 0.125 seconds. When measurements are properly aligned, salinity

spiking (and density) errors are minimized.

Note:

Real-time processing

recommendations are summarized

in Real-Time Data Corrections

in Section 4: Deploying and

Operating FastCAT.