/* Style Definitions */
mso-padding-alt:0in 5.4pt 0in 5.4pt;
mso-fareast-font-family:"Times New Roman";
mso-bidi-font-family:"Times New Roman";
This talk provides an overview of a joint effort between NOAA National Weather Service (NWS) and NOAA National Environmental Satellite Data and Information Service (NESDIS) in assessing the potential of the Self-calibrating Multivariate Precipitation Retrieval (SCaMPR) QPEs for NWS flash flood and river forecast operations. This effort involves comparisons among two SCaMPR QPEs and ground-based QPEs. The first one was derived using the operational SCaMPR, and the second one from an augmented SCaMPR with TRMM Microwave Imager and Precipitation Radar data as supplemental predictands, The two SCaMPR QPEs were assessed along with gridded gauge-only QPEs over the period of 2000-2007 for a set of basins in Texas with the drainage area ranging from 200 to 2000 km2.
The assessment was first performed through comparison of mean area rainfall time series derived from SCaMPR and gauge-only QPEs with the operational multi-sensor QPEs produced at the West Gulf River Forecast Center (WGRFC) of NWS. Our analyses on an hourly scale indicate that TRMM ingest tends to mitigate the overall low bias in baseline SCaMPR QPEs, and slightly improves the correlation and the False Alarm Ratio scores. These improvements, however, come at the expense of lower detection rate of higher rainfall amounts. Next, a set of hydrologic experiments was executed, where the Adjoint-Based Optimizer (AB-Opt) was used to estimate long-term bias in each QPE and model parameters prior to the model simulation. The resultant runoff simulations from NWSRFS based on each QPE were also compared. The results from these experiments point to some potential for the use of SCaMPR QPEs as forcing input to lumped hydrologic models, though gauge-only QPE, despite its lower correlation with the operational QPEs on an hourly basis, is more advantageous in capturing the streamflow dynamics at the prescribed gauge density.