The analytic expression relating GOES-8/9 channel-3 water vapor radiance to relative humidity is combined with National Center for Environmental Prediction (NCEP) Eta model temperature grids to produce a high horizontal resolution (8km) brightness temperature product that correlates very well with measured upper-tropospheric specific humidity. The main advantage of the product, which can be produced in near real-time, is that it allows spatial and temporal intercomparison of moisture content across an image - a feature that the standard water vapor channel lacks. It is demonstrated that much of the moisture information in GOES-8/9 channel 3 is "hidden" by strong temperature gradients and low temperatures in the upper troposphere in the original water vapor images, but is exposed in the derived water vapor product. The derived product, which also accounts for differences in satellite viewing angle, depicts variations in upper- tropospheric specific humidity with little error. We will demonstrate that the ability to remotely sense differences in moisture content increases with atmospheric dryness. This allows for resolving fine-structured dynamic features which are relevant to regional-scale forecasts. In addition, since the product can be derived globally, it has a significant potential to be used for diagnosing the large scale distribution of moisture and its relevance to climatic variations. The effectiveness of the product as a dynamic tracer is shown in a case study of a tropopause depression event over the upper-Midwest and Canada in April 1996, in which dry air of stratospheric origin is exchanged into the troposphere.