Atmospheric water vapour plays a key role in the climatology of the Earth. It has traditionally been measured using radiosondes for reasons of instrumental simplicity but these offer limited opportunities for spatial and continuous measurements of dynamic water vapour changes over large areas of the Earth's atmosphere. Efforts have recently turned to using satellite remote sensing instruments with different spectral and spatial capabilities to derive measurements of total water vapour content in atmospheric columns or simply precipitable water. The merging of remote sensing data with different spectral and spatial capabilities can result in large biases when independent measurements are not nested correctly to produce the final product. Consequently, such merging of data must take into account the intrinsic time dynamics of measured parameters. In this paper, the impact of atmospheric water vapour dynamics on the merging of satellite-based retrieval of precipitable water estimates is investigated by comparing independent measurements obtained at different spatial resolutions from the High Resolution Infrared Sounder (HIRS) and the Advanced Very High Resolution Radiometer (AVHRR). Correlations are used to infer optimal merging parameters depending on the observational conditions. The authors conclude that the merging technique reproduces HIRS-based retrievals in cloud-free and partly cloudy locations from AVHRR soundings.