There is a lot of natural variability in water vapour - geographicallly, and over a year, with seasons, with hot and cold days, cloudy, over desert, over sea, with ENSO variability etc.
All of that is understood and not ignored. To determine long term trends of increased water vapour due to atmospheric warming caused by greenhouse gasses, you need to filter out that natural variability.
