The Beckman oxygen sensor was used during cruises 13 through 22. Due to the sensor’s shortcomings explained before (Sect. 2.1), we viewed the oxygen sensor data as being useful primarily to assist in quality controlling water sample oxygens, and for interpolating between the discrete samples.

     Because the behavior of the Beckman sen­sor in the top 100 m or so appears to be poor, the shape of the O2 profile often did not resemble the distribution suggested by the bottles.

     The continuous O2 profile from the CTD was also used for screening the bottle data. It is our experience that there is considerable finestructure in O2, not always correlated with T-S finestruc­ture. This is partly due to the biological processes which make O2 a nonconservative tracer. With­out the continuous profile to reveal this structure, and depending on the distribution of bottle sam­ples, one might reject bottle oxygen values that are accurate. Thus, even though the sensor has some undesirable characteristics, it still provides useful information on the variability of O2 on scales smaller than or comparable to the bottle spacing.

     CTD O2 calibration was performed as explained before (Section 3). Two sets of parameters were obtained per cruise, corresponding to the casts of Station 1 (Kahe Pt.) and Station 2 (ALOHA). For a better fit, data deeper than 1000 m from station 2 were assigned extra weight given that only one deep cast was obtained per cruise. This weighting was done by duplicating the data below 1000 m as many times as shal­low (< 1000 m) casts were available for the calibration.

     In some of the cruises, t was kept constant (t = 20, nominal in cruises 16, 17 and 18, station 2), rather than allowing it to be a negative value. This did not noticeably affect the results of the calibrations.

     In cruises 19, 20, and 22, the CTD traces in the upper 100 dbar showed large deviations from the O2 bottle data, leading to a lack of fit of the whole cast during the calibration. The fit largely improved after neglecting the upper 100 dbar. The CTD O2 was flagged as bad or suspicious in the upper 100 dbar of these casts (see Figure 4.2). The quality of the O2 sensor calibration can be seen in table 4.2.

     Figure 4.2 shows the CTD oxygen data quality for every 2-dbar bin in each cast at Station ALOHA for cruises 13 through 22. Most of the casts from HOT-13 and all casts from HOT-15 did not have an oxygen sensor.

Figure 4.2. CTD oxygen data quality per cast (vertical bars) and per 2-dbar bin at Station ALOHA during HOT cruises 13 through 22. The tick marks along the x-axis separate different casts. Casts from the same cruise are separated by vertical black lines.

Table 4.2: CTD-Bottle O2 (mmol kg-1) Comparison for Each Cruise

 

Station 1, Kahe Point

Station 2, ALOHA

0 < P < 1100

0 < P < 4700

500 < P < 4700

Cruise

Mean

St. dev

Mean

St. dev

Mean

St. dev

HOT-13

*

*

0.07

1.46

0.19

1.45

HOT-14

-0.01

2.52

0.04

1.33

0.23

1.46

HOT-15

**

**

**

**

**

**

HOT-16

0.01

2.27

0.38

3.87

0.09

1.95

HOT-17

-0.12

3.21

-0.47

3.49

-0.75

2.36

HOT-18

-0.25

2.87

-1.12

4.09

-0.88

3.16

HOT-19

0.00

2.91

-1.51

4.35

-1.15

4.22

HOT-20

-0.03

2.68

***

***

***

***

HOT-22

-0.20

1.55

0.29

3.43

-0.07

3.55

* No 02 sensor during this station

** No 02 sensor on CTD during this cruise

*** CTD lost during Station 2, cast 1

Back to Table of Contents

Previous Next