Making Satellite Precipitation Data Work for the Developing World

In an era of declining ground-based networks for measurement of precipitation, satellite precipitation data, that is now routinely available in increasing frequency and spatial coverage, represents an interesting paradox. Specifically, effective assessment frameworks and error metrics for satellite precipitation data must be developed for advancing the utility of satellite data for global applications. While there seems to be a concerted effort by the community to evaluate satellite precipitation data, there does not seem to be similar efforts to resolve the paradoxical issue of balancing the need for global uncertainty information and the stark lack of global GV datasets for doing so. In this article, we present one practical approach to estimating satellite precipitation uncertainty that is not dependent on the notion of ground validation (GV) data. By using input that is more readily available around the globe (i.e., satellite data and geophysical features of terrain, climate and seasons), the approach can potentially advance applications as it allows a coherent way to merge available satellite precipitation data products to a more superior state, particularly for hydrologic applications. We provide an assessment of how the approach works in various regions of the developing world as a way to encourage the community to further the development of such ideas and provide end-users with a practical decision-making tool.