Quantitative precipitation estimation (QPE) remains as one of the principal observation objectives for weather researchers and forecasters. Traditionally reflectivity has been used to relate radar measurements to rain rate (e.g., Fulton et al. 1998; Willie et al. 2017). The purpose of this paper is to discuss the development of a regional dual polarization QPE process known as the RAdar Multi-Sensor QPE (RAMS QPE). This process applies the dual polarization radar rain rate estimation algorithms established by Chen et al. (2017) into a QPE system that operates over a user-defined domain. The methodologies that are used to ingest and combine individual radar scans, and then merge them into a mosaic are described. The implementation and evaluation is executed over a domain that extends over Northern California centered near San Francisco and the bay area. The events for this study occurred during the month of January and February of 2017 that include several flooding events. The QPE precipitation fields evaluated in this analysis are derived from the dual polarization radar data obtained from the local National Weather Service (NWS) WSR-88DP (i.e., KBBX, KDAX, KMUX) radars and from the NWS QPE radar-only products.
Methodology
The notion of merging measurements from radar, gauge and other sensing instruments into QPE products has been established by the National Severe Storms Laboratory (NSSL) called Multi Radar-Multi Sensor (MRMS) (Zhang et al. 2011). This MRMS system is designed to ingest NEXRAD data, such that the domain extends over the entire conterminous United States. The application of this QPE system for regional applications is possible, however the configuration is tedious and computational resources are substantial. The proposed regional polarimetric, RAMS QPE, scheme looks to simplify the usage of applying a QPE system that operates on a smaller domain and where the focus is to generate rainfall from any surrounding radars that provide dual-polarization measurements.
The flexibility in radar input allows for the use of other regional or mobile radars that meet the requirements of polarimetric capability. The radars used within the domain are individually processed to generate QPE using the dual polarization techniques developed by Colorado State University (Chen et al. 2017). Once the precipitation estimation maps are generated for each radar sweep, single radar gridding is performed to combine the single radar scans. The set of individual radar sweeps that occur within a particular timeframe are combined to create a Single Radar Hybrid Scan of Dual Polarization (HSDP). To accomplish this, the radar radials are re-gridded from intrinsic radar polar coordinates into a rectangular coordinate system with a precision of 0.01 deg by 0.01 deg in latitude-longitude coordinates. Using the single radar HSDP maps, which extend over the region of interest, these are then merged together using a simple combinational technique to create a mosaicked QPE. The intent of the merge is to preserve the lowest scans in order to better estimate rainfall near the ground and below the melting layer.
The RAMS QPE system is also capable of generating an independent gauge QPE that employs an inverse distance weighting (IDW) scheme and is based upon the method of Simanton and Osborn (1980), where fi = gauge value, b = power parameter, d = distance from interpolation point to gauge, and i is the gauge number. The equations are applied using b=2 and a 2 km radius of influence is used.
EVALUATION
The RAMS QPE radar-only and gauge-only product will be evaluated over the Northern California domain shown in the Figure. The validation gauge data will come from a set of 54 gauges that are located within the domain and are independent of the MRMS radar-only and gauge-only products. The performance will be determined by comparing the RAMS QPE grid points with this validation gauge data set. The MRMS radar-only product will also be evaluated over the same region as to see how well it accomplishes the same task.
References
Chen, Haonan, V. Chandrasekar, and R. Bechini, 2017: An Improved Dual-Polarization Radar Rainfall
Algorithm (DROPS2.0): Application in NASA IFloodS Field Campaign. Journal of Hydrometeorology, 18, 917-937.
Fulton, R.A., J.P. Breidenbach, D. Seo, D.A. Miller, and T. O’Bannon, 1998: The WSR-88D Rainfall
Algorithm. Wea. Forecasting, 13, 377-39.5
Simanton, J. R., and H. B. Osborn, 1980: Reciprocal-distance estimate of point rainfall. Journal of the
Hydraulics Division, 106 (7), 1242-1246.
Willie, D., H. Chen, V. Chandrasekar, R. Cifelli, and et al., 2017: Evaluation of Multisensor
Quantitative Precipitation Estimation in Russian River Basin. Journal of Hydrologic Engineering, 22(5), doi:10.1061/(ASCE)HE.1943-5584.0001422.
Zhang, J., and Coauthors, 2011: National Mosaic and Multi-Sensor QPE (NMQ) System: Description,
Results, and Future Plans. Bull. Amer. Meteor. Soc., 92, 1321-1338.
- Indicates paper has been withdrawn from meeting
- Indicates an Award Winner