Comparison of Extremes in Stage IV, TRMM 3B42v7 and MERRA2 Observation Corrected Precipitation

Quantitative Precipitation Estimation (QPE) is a difficult challenge, yet is critically important to weather forecasting and climate research. Many methods can be used to arrive at the estimation of how much rain has fallen, including ground based gauges, radar estimations, satellite data, reanalysis, and combinations of these methods. Stage IV, MERRA2 (Modern Era Research and Reanalysis Version 2) and TRMM (Tropical Rainfall Measurement Mission) are three prominent datasets for precipitation research, which use very different input observations. As a reanalysis, MERRA2 is at a coarser resolution (0.5 x 0.625 degrees) than TRMM (0.25 x 0.25 degrees), a microwave and radar satellite product and Stage IV (4km x 4km), a combination radar and gauge dataset. Especially when producing QPE for extreme events, these products contain uncertainties relating to both resolution and the uncertainty in observational datasets used as input. Understanding of each dataset’s intrinsic biases aids in contextualizing the research done with these datasets. Analyzing when an event is extreme in the context of one dataset but not the other gives an indication of what meteorological conditions are like when these datasets disagree most with each other. This in turn allows for an understanding of the strengths and limitations of that dataset.