Oxygen optode measurements on floats and gliders suffer from a
slow time response and various sources of drift in the calibration
coefficients. Based on two dual-

While oceanic oxygen measurements of the last century were mostly based on
Winkler titrations of discrete water samples or profiles acquired with
CTD-mounted electrochemical oxygen sensors, such observations rely
increasingly on

Apart from cost effectiveness, the main reasons for this are improved sensor
performance and characterization with respect to temperature and

It is therefore timely and useful to revisit the foundations of both pumped
and unpumped

This study utilizes data from two dual-

Only the unpumped Aanderaa 4330 optode, cable-mounted in an elevated position
on the float's top cap, is capable of in-air measurements, and the float's
firmware was adjusted accordingly (see

Shipboard CTD–

Since the SBE63 optode's sensing foil is contained inside the pumped path of
the float's CTD, its water flow is set by the pump speed of the float CTD,
i.e., ca. 600 mL min

Using the faster SBE63 optode

define the flow regime for the pumped SBE63 optode, i.e.,

translate the flow regime into a response time constant

obtain the mean of consecutive data points of the unbiased

reconstruct the unbiased

The analysis of the Aanderaa 4330 optode flow regime behavior takes the
opposite approach by filtering or “delaying” the reference

estimate a boundary layer thickness

translate

translate

apply the recursive filter to the mean of consecutive data points of the unbiased

compare the filtered

Boundary layer thickness

Overall, the resulting boundary layer thickness

This flow regime characterization can now be used to reconstruct a time
response-unbiased

Profiles of ascent speed, density, SBE63 optode

An example of such a reconstruction is shown in
Fig.

Median profiles (thick lines) with 10th/90th percentiles (thin
lines) for floats 6900889 (green) and 6900890 (black):

As expected, the time response correction for the pumped SBE63 optode with
small response times changes the

This demonstrates that a time response correction is both useful for removing
sensor artifacts (Fig.

Interestingly, a similar analysis of the limited dataset of the third float,
WMO ID 6900891, yields apparent boundary layer thicknesses

The feasibility in calibrating

Time series (top) as well as daytime dependence (bottom) of the
oxygen correction slope

In contrast to

Same as Fig.

One can postulate that the daytime bias is related to solar heating of the
sensor's sensing foil and thus the solar elevation angle

For float 6900889,

To evaluate the sensitivity to the atmospheric reanalysis product, air

Simultaneous fit of “carry-over” slope

95

Based on regular in-air measurements, the Aanderaa 4330 optodes show a
significant drift of

Percent difference between

Since the SBE63 optode on floats can not measure oxygen in-air, a direct
comparison with in-air measurements is not possible. However, it can be done
indirectly via the difference between the 4330 and SBE63 optodes
(Fig.

Moreover, at both depths, there seems to be a conditioning effect during the
first half-year (

Linear fit slopes of the sensor difference (using only data after 180 days)
for the two floats are indistinguishable from zero within

A slow

In-air measurements at every surfacing of Argo-

Based on regular in-air measurements over 38 and 25 months, respectively, we
were able to detect a significant in situ drift in both 4330 optodes of

Given the findings of several studies on both optode “storage” drift and in
situ drift, we strongly propose that every Biogeochemical-Argo float be
deployed with in situ

To our knowledge, the feasibility of in-air measurements on gliders still needs to be shown. Given the generally shorter deployment periods and, more importantly, the possibility of careful pre- and post-deployment calibration/drift characterization (e.g., by optode in-air measurements under controlled conditions on deck), it might not be necessary to have in situ in-air measurements on gliders.

The float data are made freely available by the International Argo Program
(Argo, 2000) and the national programs that contribute to it
(

Here, we reproduce the essential equations from

The

For the reconstruction, Eq. (

For a pumped optode, the relation between flow

The boundary layer thickness

The authors want to thank Patrice Brehmer (IRD, Dakar, Senegal) as well as
the captain and crew of M/V