资源环境科技发展态势分析平台

This study demonstrates the potential use of the NASA's Global Precipitation Measurement (GPM) Dual-frequency Precipitation Radar (DPR) to examine ground radar (GR) miscalibration and other uncertainty sources (e.g., partial beam blockage). We acquired the GPM Ground Validation System Validation Network reflectivity matchups between the DPR and three GRs (two in Iowa and one in South Dakota) for 2014-2017. We then refined the matching parameters (e.g., time separation) to reduce uncertainty in the matchup samples by analyzing the sensitivity of the matchup statistical properties to changes in these parameters. To reconcile the same observables (i.e., reflectivity) with different observational properties among the space- and ground-based radars, we developed a statistically integrated framework using inter-comparisons of them all with a Monte Carlo simulation. This method verifies the absolute calibration bias estimated from the refined DPR-GR matchups using relative calibration biases between GRs. We found that taking samples with a narrow temporal gap, estimated by actual measurement time of the DPR and GRs, can significantly reduce sample variability. Through inter-comparisons among the DPR and GRs, we observed that reflectivity differences among GRs in a similar environment (e.g., climatology and geography) are likely to be affected primarily by the calibration mismatch. In this case, the inter-comparison results demonstrated good agreement, and we inferred that the differences can be mitigated by calibration bias correction against the DPR. On the other hand, when the disagreement level of the inter-comparison results is significant, the authors found that other factors, such as partial beam blockage even in relatively plain regions, are more dominant than the calibration bias. In fact, the partial beam blockage effects can manifest themselves as a seasonal pattern in the GR inter-comparison results.