Search Results

You are looking at 1 - 10 of 12 items for

  • Author or Editor: David Kitzmiller x
  • All content x
Clear All Modify Search
Yu Zhang, Seann Reed, and David Kitzmiller

Abstract

This paper presents methodologies for mitigating temporally inconsistent biases in National Weather Service (NWS) real-time multisensor quantitative precipitation estimates (MQPEs) through rain gauge–based readjustments, and examines their effects on streamflow simulations. In this study, archived MQPEs over 1997–2006 for the Middle Atlantic River Forecast Center (MARFC) area of responsibility were readjusted at monthly and daily scales using two gridded gauge products. The original and readjusted MQPEs were applied as forcing to the NWS Distributed Hydrologic Model for 12 catchments in the domain of MARFC. The resultant hourly streamflow simulations were compared for two subperiods divided along November 2003, when a software error that gave rise to a low bias in MQPEs was fixed. It was found that readjustment at either time scale improved the consistency in the bias in streamflow simulations. For the earlier period, independent monthly and daily readjustments considerably improved the streamflow simulations for most basins as judged by bias and correlation. By contrast, for the later period the effects were mixed across basins. It was also found that 1) readjustments tended to be more effective in the cool rather than warm season, 2) refining the readjustment resolution to daily had mixed effects on streamflow simulations, and 3) at the daily scale, redistributing gauge rainfall is beneficial for periods with substantial missing MQPEs.

Full access
David H. Kitzmiller and Wayne E. Mcgovern

Abstract

Wind profiler, rawinsonde, and surface observations of the atmosphere over northeastern Colorado during the morning hours on 44 days were compared to the severity of subsequent thunderstorm activity. On half of thes days, large hail (diameter ≥2 cm) was observed over the region, while on the other half, only thunderstorms with no large hail or other severe local storm phenomena were reported. Statistical comparisons revealed that the wind speed near 8 km above ground level (AGL), the southerly wind component between 2.0 and 2.5 km AGL, and a thermal advection index computed from the degree of wind veering in the 1.5–2.5-km layer, were all significantly greater on the large-hail days than on the nonsevere weather days. Concurrently available rawinsonde observations did not detect some of these differences as clearly as did the profiler observations. A screening discriminant analysis of possible predictor combinations showed that the optimum discrimination between the cases with and without large hail was given by a linear combination of 8-km wind speed from profiler measurements and positive buoyant energy from rawinsonde temperature profiles.

Full access
David H. Kitzmiller and Wayne E. McGovern

Abstract

Objective experiments have been carried out to determine which moisture and stability indices as derived from the VISSR Atmospheric Sounder (VAS) contain the greatest amount of predictive information with respect to thunderstorm and severe local storm events. In these experiments, stability and moisture parameters derived from 1700 UTC VAS retrievals were compared and correlated to storm observations made during the subsequent 2000–0000 UTC period. The amount of predictive information in these indices was also compared to that possessed by indices derived from VAS first-guess profiles, concurrently available rawinsonde measurements, and numerical model forecasts. The correlation in the form of the computed information ratio (Ic) was used as a measure of predictive power in these experiments.

It was found that precipitable water and modified versions of the classic K index which included recent surface data had the highest values of Ic for general thunderstorm occurrence. The 50-kPa gradient wind speed (derived from VAS geopotential heights) and the temperature lapse rate in the 70–50 kPa layer were the best predictors for discriminating severe local storm cases from general thunderstorm cases. The 3-h and 6-h changes in VAS stability and moisture indices were poorly correlated to severe storm occurrence. Of all the variables examined, only the 3-h and 6-h change in surface blackbody temperature appeared to be even moderately correlated to severe storms.

A series of probability forecast and verification experiments was carried out to determine if the incorporation of VAS observations might improve automated thunderstorm probability forecasts. It was found that the retrievals possessed more information with respect to thunderstorm occurrence than do their own first guess profiles, which were derived from a forecast of the 0000 UTC limited-area Fine Mesh (LFM) model. However, the VAS-derived stability at 1700 UTC appears to possess little or no additional information beyond that available from 1200 UTC rawinsonde measurements, and indices derived from 1200 UTC LFM forecasts valid near the observation period (2000–0000 UTC) have more information than the 1700 UTC VAS-based indices. Possible reasons for these findings, and their implications for future satellite operations and applied research are discussed.

Full access
Wanru Wu, David Kitzmiller, and Shaorong Wu

Abstract

This study evaluated 24-, 6-, and 1-h radar precipitation estimated from the National Mosaic and Multisensor Quantitative Precipitation Estimation System (NMQ) and the Weather Surveillance Radar-1988 Doppler (WSR-88D) Precipitation Processing System (PPS) over the conterminous United States (CONUS) for the warm season April–September 2009 and the cool season October 2009–March 2010. Precipitation gauge observations from the Automated Surface Observing System (ASOS) were used as the ground truth. Gridded StageIV multisensor precipitation estimates were applied for supplementary verification. The comparison of the two systems consisted of a series of analyses including the linear correlation coefficient (CC) and the root-mean-square error (RMSE) between the radar precipitation estimates and the gauge observations, large precipitation amount detection categorical scores, and the reliability of precipitation amount distribution. Data stratified for the 12 CONUS River Forecast Centers (RFCs) and for the cold rains events with bright-band effects were analyzed additionally. Major results are 1) the linear CC of NMQ versus ASOS are generally higher than that of PPS versus ASOS over CONUS, while the spatial variations stratified by the RFCs may switch with seasons; 2) compared to the precipitation distribution of ASOS, NMQ shows less deviation than PPS; 3) for the cold rains verified against ASOS, NMQ has higher CC and PPS has lower RMSE for 6-h and higher RMSE for 1-h cold rains; and 4) for the precipitation detection categorical scores, either NMQ or PPS can be superior, depending on the time interval and season. The verification against StageIV gridded precipitation estimates showed that NMQ consistently had higher correlations and lower biases than did PPS.

Full access
David H. Kitzmiller, Wayne E. McGovern, and Robert F. Saffle

Abstract

The WSR-88D severe weather potential (SWP) algorithm is an automated procedure for the detection of severe local storms. The algorithm identifies individual thunderstorm cells within radar imagery and, for each cell, yields an index proportional to the probability that the cell will shortly produce damaging surface winds, large hail, or tornadoes. This index is a statistically derived function of the storm's maximum vertically integrated liquid (VIL) and horizontal areal extent. The correlation between these storm characteristics and severe weather occurrence was first documented in the 1970s. Several National Weather Service field offices in the central plains and Northeast regions of the United States have successfully used VIL as a discriminator between severe and nonsevere thunderstorms.

This paper describes the observational data and statistical methodology employed in the development of the SWP algorithm, and regional and seasonal variations in the SWP/severe weather relationship. The expected operational performance of the algorithm, in terms of probability of detection and false alarm ratio, is also documented.

Full access
Jay P. Breidenbach, David H. Kitzmiller, Wayne E. McGovern, and Robert E. Saffle

Abstract

The operational WSR-88D Severe Weather Potential (SWP) algorithm is an automated nowcasting procedure aimed at providing guidance in the detection of severe local storms. It yields a numerical index proportional to the probability that an individual storm cell is producing, or will shortly produce, large hail, damaging surface winds, or tornadoes.

Currently, the SWP algorithm consists of a statistically derived function of the cell's maximum vertically integrated liquid and horizontal areal extent. In an attempt to refine the algorithm, a wide variety of new statistical predictors of severe weather have been derived from volumetric reflectivity observations. Experimental second-generation SWP equations incorporating these new predictors were evaluated and their skill was compared to that of the operational SWP algorithm.

Those predictors that parameterize the magnitude of the reflectivity in the middle and upper portions of convective storms were found to have the most diagnostic information with respect to severe weather. Some of these predictors rely only on reflectivity above 15 000 ft (4572 m) and thus could be applied to storms beyond the current algorithm's range of 230 km. The skill of the second-generation equations within 230 km was found to be comparable to that of the current algorithm.

Full access
Haksu Lee, Yu Zhang, Dong-Jun Seo, Robert J. Kuligowski, David Kitzmiller, and Robert Corby

Abstract

This study examines the utility of satellite-based quantitative precipitation estimates (QPEs) from the Self-Calibrating Multivariate Precipitation Retrieval (SCaMPR) algorithm for hydrologic prediction. In this work, two sets of SCaMPR QPEs, one without and the other with Tropical Rainfall Measurement Mission (TRMM) version 6 data integrated, were used as input forcing to the lumped National Weather Service hydrologic model to retrospectively generate flow simulations for 10 Texas catchments over 2000–07. The year 2000 was used for the model spinup, 2001–04 for calibration, and 2005–07 for validation. The results were validated using observed streamflow alongside similar simulations obtained using interpolated gauge QPEs with varying gauge network densities, and still others using the operational radar–gauge multisensor product (MAPX). The focus of the evaluation was on the high-flow events. A number of factors that could impact the relative utility of SCaMPR satellite QPE and gauge-only analysis (GMOSAIC) for flood prediction were examined, namely, 1) the incremental impacts of TRMM version 6 data ingest, 2) gauge density, 3) effects of calibration approaches, and 4) basin properties. Results indicate that ground-sensor-based QPEs in a broad sense outperform SCaMPR QPEs, while SCaMPR QPEs are competitive in a minority of catchments. TRMM ingest helped substantially improve the SCaMPR QPE–based simulation results. Change in calibration forcing, that is, calibrating the model using individual QPEs rather than the MAPX (the most accurate QPE), yielded overall improvements to the simulation accuracy but did not change the relative performance of the QPEs.

Full access
Yu Zhang, Dong-Jun Seo, David Kitzmiller, Haksu Lee, Robert J. Kuligowski, Dongsoo Kim, and Chandra R. Kondragunta

Abstract

This paper assesses the accuracy of satellite quantitative precipitation estimates (QPEs) from two versions of the Self-Calibrating Multivariate Precipitation Retrieval (SCaMPR) algorithm relative to that of gridded gauge-only QPEs. The second version of SCaMPR uses the QPEs from Tropical Rainfall Measuring Mission (TRMM) Precipitation Radar and Microwave Imager as predictands whereas the first version does not. The assessments were conducted for 22 catchments in Texas and Louisiana against National Weather Service operational multisensor QPE. Particular attention was given to the density below which SCaMPR QPEs outperform gauge-only QPEs and effects of TRMM ingest. Analyses indicate that SCaMPR QPEs can be competitive in terms of correlation and CSI against sparse gauge networks (with less than one gauge per 3200–12 000 km2) and over 1–3-h scale, but their relative strengths diminish with temporal aggregation. In addition, the major advantage of SCaMPR QPEs is its relatively low false alarm rates, whereas gauge-only QPEs exhibit better skill in detecting rainfall—though the detection skill of SCaMPR QPEs tends to improve at higher rainfall thresholds. Moreover, it was found that ingesting TRMM QPEs help mitigate the positive overall bias in SCaMPR QPEs, and improve the detection of moderate–heavy and particularly wintertime precipitation. Yet, it also tends to elevate the false alarm rate, and its impacts on detection rates can be slightly negative for summertime storms. The implications for adoption of TRMM and Global Precipitation Measurement (GPM) QPEs for NWS operations are discussed.

Full access
David Kitzmiller, Suzanne Van Cooten, Feng Ding, Kenneth Howard, Carrie Langston, Jian Zhang, Heather Moser, Yu Zhang, Jonathan J. Gourley, Dongsoo Kim, and David Riley

Abstract

This study investigates evolving methodologies for radar and merged gauge–radar quantitative precipitation estimation (QPE) to determine their influence on the flow predictions of a distributed hydrologic model. These methods include the National Mosaic and QPE algorithm package (NMQ), under development at the National Severe Storms Laboratory (NSSL), and the Multisensor Precipitation Estimator (MPE) and High-Resolution Precipitation Estimator (HPE) suites currently operational at National Weather Service (NWS) field offices. The goal of the study is to determine which combination of algorithm features offers the greatest benefit toward operational hydrologic forecasting. These features include automated radar quality control, automated ZR selection, brightband identification, bias correction, multiple radar data compositing, and gauge–radar merging, which all differ between NMQ and MPE–HPE. To examine the spatial and temporal characteristics of the precipitation fields produced by each of the QPE methodologies, high-resolution (4 km and hourly) gridded precipitation estimates were derived by each algorithm suite for three major precipitation events between 2003 and 2006 over subcatchments within the Tar–Pamlico River basin of North Carolina. The results indicate that the NMQ radar-only algorithm suite consistently yielded closer agreement with reference rain gauge reports than the corresponding HPE radar-only estimates did. Similarly, the NMQ radar-only QPE input generally yielded hydrologic simulations that were closer to observations at multiple stream gauging points. These findings indicate that the combination of ZR selection and freezing-level identification algorithms within NMQ, but not incorporated within MPE and HPE, would have an appreciable positive impact on hydrologic simulations. There were relatively small differences between NMQ and HPE gauge–radar estimates in terms of accuracy and impacts on hydrologic simulations, most likely due to the large influence of the input rain gauge information.

Full access
Jian Zhang, Kenneth Howard, Carrie Langston, Steve Vasiloff, Brian Kaney, Ami Arthur, Suzanne Van Cooten, Kevin Kelleher, David Kitzmiller, Feng Ding, Dong-Jun Seo, Ernie Wells, and Chuck Dempsey

The National Mosaic and Multi-sensor QPE (Quantitative Precipitation Estimation), or “NMQ”, system was initially developed from a joint initiative between the National Oceanic and Atmospheric Administration's National Severe Storms Laboratory, the Federal Aviation Administration's Aviation Weather Research Program, and the Salt River Project. Further development has continued with additional support from the National Weather Service (NWS) Office of Hydrologic Development, the NWS Office of Climate, Water, and Weather Services, and the Central Weather Bureau of Taiwan. The objectives of NMQ research and development (R&D) are 1) to develop a hydrometeorological platform for assimilating different observational networks toward creating high spatial and temporal resolution multisensor QPEs for f lood warnings and water resource management and 2) to develop a seamless high-resolution national 3D grid of radar reflectivity for severe weather detection, data assimilation, numerical weather prediction model verification, and aviation product development.

Through about ten years of R&D, a real-time NMQ system has been implemented (http://nmq.ou.edu). Since June 2006, the system has been generating high-resolution 3D reflectivity mosaic grids (31 vertical levels) and a suite of severe weather and QPE products in real-time for the conterminous United States at a 1-km horizontal resolution and 2.5 minute update cycle. The experimental products are provided in real-time to end users ranging from government agencies, universities, research institutes, and the private sector and have been utilized in various meteorological, aviation, and hydrological applications. Further, a number of operational QPE products generated from different sensors (radar, gauge, satellite) and by human experts are ingested in the NMQ system and the experimental products are evaluated against the operational products as well as independent gauge observations in real time.

The NMQ is a fully automated system. It facilitates systematic evaluations and advances of hydrometeorological sciences and technologies in a real-time environment and serves as a test bed for rapid science-to-operation infusions. This paper describes scientific components of the NMQ system and presents initial evaluation results and future development plans of the system.

Full access