A new method to correct the ECC ozone sonde time response and its implications for “background current” and pump efficiency

Abstract. The Electrochemical Concentration Cell (ECC) ozonesonde has been the main instrument for in situ profiling of ozone worldwide; yet, some details of its operation, which contribute to the ozone uncertainty budget, are not well understood. Here, we investigate the time response of the chemical reactions inside the ECC and how corrections can be used to remove some systematic biases. The analysis is based on the understanding that two reaction pathways involving ozone occur inside the ECC that generate electrical currents on two very different time scales. A slow reaction pathway involving the buffer with a time constant of about 25 min can be interpreted as what has conventionally been considered the “background current”. This contribution can be calculated and removed from the measured current instead of the “background current”. The remaining fast reaction pathway with a time constant of about 20 s is due the conversion of iodide to molecular iodine and the generation of two free electrons per ozone molecule. Here we provide an algorithm to calculate and remove the contribution of the slow reaction pathway and to correct for the time lag of the faster reaction pathway. This processing algorithm has been applied to ozonesonde profiles at Costa Rica and during the Central Equatorial Pacific Experiment (CEPEX) and to laboratory experiments evaluating the performance of ECC ozonesondes. At Costa Rica, where a 1 % KI, 1/10th buffer solution is used, there is no change in the derived total ozone column; however, in the upper troposphere and lower stratosphere, average reported ozone concentrations increase by up to 7 % and above 30 km decrease by up to 7 %. During CEPEX, where a 1 % KI, full buffer solution was used, ozone concentrations are increased mostly in the upper troposphere with no change near to the top of the profile. In the laboratory measurements, the processing algorithms have been applied to measurements using all current sensing solutions and using only the stronger pump efficiency correction reported by Johnson et al. (2002), which improves the time response of the ECCs and removes some biases relative to the reference instruments. In the surface layer, the correction algorithm shows that ECC ozonesonde measurements are influenced by the operational procedures prior to launch and that typical gradients above the surface layer may be steeper than originally reported.