The average large gas bubble size in slurry bubble columns is estimated using a spectral analysis method applied to measured pressure time series. A pressure time series measured in a bubble column consists of local pressure fluctuations and global pressure fluctuations. The local pressure fluctuations arise from the liquid velocity fluctuations induced by the large gas bubbles and by the changes in gas holdup. The standard deviation of these local pressure fluctuations is a measure of the average large bubble size. The coherence between the pressure time series measured at the sparger and at any other location in the column is used to separate the local pressure fluctuations from the global pressure fluctuations. The global pressure fluctuations are measured instantaneously throughout the column and form the coherent part of the pressure time series. The local pressure fluctuations, which are absent at the sparger, form the incoherent part of the pressure time series. In a 2-D bubble column, a clear correlation is demonstrated between the incoherent standard deviation of the pressure time series and the average large gas bubble size obtained from video imaging. A good agreement is also found between this correlation and the 3-D model proposed by Krishna et al. for predicting the average large bubble size in 3-D (slurry) bubble columns. (c) 2005 American Institute of Chemical Engineers.