In recent years several methods have been developed that derive total tropospheric 03 columns from satellite measurements. However, one issue that has not been paid much attention to is the interpretation of (extratropical) total tropospheric O<inf>3</inf> columns. Different processes contribute to the total tropospheric 03 column: stratosphere-troposphere exchange, tropospheric O <inf>3</inf> production and loss, transport and the height of the tropopause. Each process contributes differently to the total tropospheric O<inf>3</inf> column variability depending on season, geographical location, and altitude. This paper investigates the contribution of these different processes on total tropospheric O<inf>3</inf> column variability using a chemistry-climate model simulation of tropospheric 03, and reflects on the implications for total tropospheric 03 column measurements. On the basis of tropospheric 03 column (satellite) measurements and without other sources of information (e.g., model simulations, observations of other trace species) it is not possible to determine the separate contributions by the aforementioned processes to the extratropical total tropospheric 03 column variability. Furthermore, typical extratropical synoptic-scale (daily) total tropospheric 03 column variability is of the order of 10 DU (1-σ value), implying the errors in (satellite) measurements should be of the order of magnitude at maximum (~10 DU) for daily measurements. For tropical total tropospheric O<inf>3</inf> column (satellite) measurements the requirements are less stringent because the most important variability occurs on seasonal timescales. Errors in tropical total tropospheric 03 column (TTOC) measurements should be of the order of 5 DU for monthly means. Copyright 2005 by the American Geophysical Union.