Previous efforts at climate signal detection using the space-time optimal filter approach have been limited to the time evolution of surface temperature data. This limitation arises from the lack of long-term data sets of temperature which have sufficient data density in both the horizontal and vertical dimensions of the atmosphere to be useful in such an approach. Future data sets from satellites should provide a wealth of such data. To investigate the potential improvement in climate signal detection when sufficient vertical information becomes available, data from a coupled ocean-atmosphere climate model is used. For a transient simulation of the doubling of carbon dioxide in the atmosphere, it is found that the signal-to-noise ratio is increased by a factor of 2.28 when temperature data from the tropical upper troposphere and stratosphere are used, as compared with using only the tropical surface temperature data. When only tropospheric data are used, the signal-to-noise ratio is increased by a factor of 1.12. ¿ 1999 American Geophysical Union