Search Results

You are looking at 1 - 2 of 2 items for :

  • Author or Editor: M. J. Atkinson x
  • Journal of Atmospheric and Oceanic Technology x
  • Refine by Access: All Content x
Clear All Modify Search
F. I. M. Thomas and M. J. Atkinson


A new type of oxygen sensor for use on oceanic profiling CTDs has a response time that ranges from 0.4 to 0.7 s, depending on temperature. This oxygen (O2) sensor was calibrated in the field against O2 concentrations that were determined from Winkler titrations of Niskin-bottle water samples. Calibration coefficients were obtained from single casts and composite casts. When casts were calibrated separately, the mean error in predicting O2 concentrations was 1.22 ± 0.72 µM O2, n = 19 casts, and was not significantly different from the error between Niskin-bottle samples, which was 1.01 ± 2.33 µm O2, n = 33 triplicate or duplicate bottles. When casts wore combined into a composite cast for an entire cruise, however, the mean error in the calibrations increased to 5.03 ± 4.06 µm O2, n = 19 casts. These results also indicate that only three Niskin bottles on a cast are required to obtain an error of 1.0 µm in predicted oxygen. For routine use, we recommend calibrating each cast separately using either three or four measured O2 concentrations to obtain calibration coefficients.

Full access
M. J. Atkinson, F. I. M. Thomas, and N. Larson


To measure the effects of pressure on the output of a membrane oxygen sensor and a nonmembrane oxygen sensor, the authors pressure cycled a CTD sensor package in a laboratory pressure facility. The CTD sensor package was cycled from 30 to 6800 db over a range of temperatures from 2° to 38°C. Pressure decreased the output of the membrane sensor and increased the output of the microhole sensor. The pressure terms for both types of oxygen sensors were affected by temperature. The effect of pressure on both types of sensors can be quantified as exp (VP/RT), where V is a coefficient (cm3 mol−1), P is decibars, R′ is the gas constant (831.47 cm3 mol−1 db K−1), and T is kelvins. As water gets colder, V for both sensors increases. For temperatures less than 21°C, V for the membrane sensor is −33.7±0.54 cm3 mol−1, and V for the microbole sensor is 0.29±0.31 cm3 mol−1. The V's for calibrations of four oceanic casts had larger ranges than the laboratory experimental data: −27.6 to −34.9 cm3 mol−1 for the membrane sensor, and −0.4 to −2.9 cm3 mol−1 for the microhole sensor. At 10°C, increasing pressure to depths of 5000 m decreases current output of a membrane sensor approximately 50% and increases output of a microhole sensor about 0.6%. For field calibrations, the authors recommend using a constant V obtained by iterations of linear fits. The use of a pressure term with the form exp(VP/RT) appears to improve field calibrations of membrane oxygen sensors.

Full access