Abstract

A hybrid parallel scheme for the ensemble square root filter (EnSRF) suitable for parallel assimilation of multiscale observations, including those from dense observational networks such as those of radar, is developed based on the domain decomposition strategy. The scheme handles internode communication through a message passing interface (MPI) and the communication within shared-memory nodes via Open Multiprocessing (OpenMP) threads. It also supports pure MPI and pure OpenMP modes. The parallel framework can accommodate high-volume remote-sensed radar (or satellite) observations as well as conventional observations that usually have larger covariance localization radii.

The performance of the parallel algorithm has been tested with simulated and real radar data. The parallel program shows good scalability in pure MPI and hybrid MPI–OpenMP modes, while pure OpenMP runs exhibit limited scalability on a symmetric shared-memory system. It is found that in MPI mode, better parallel performance is achieved with domain decomposition configurations in which the leading dimension of the state variable arrays is larger, because this configuration allows for more efficient memory access. Given a fixed amount of computing resources, the hybrid parallel mode is preferred to pure MPI mode on supercomputers with nodes containing shared-memory cores. The overall performance is also affected by factors such as the cache size, memory bandwidth, and the networking topology. Tests with a real data case with a large number of radars confirm that the parallel data assimilation can be done on a multicore supercomputer with a significant speedup compared to the serial data assimilation algorithm.

Abstract

NOAA’s National Severe Storms Laboratory is actively developing phased-array radar (PAR) technology, a potential next-generation weather radar, to replace the current operational WSR-88D radars. One unique feature of PAR is its rapid scanning capability, which is at least 4–5 times faster than the scanning rate of WSR-88D. To explore the impact of such high-frequency PAR observations compared with traditional WSR-88D on severe weather forecasting, several storm-scale data assimilation and forecast experiments are conducted. Reflectivity and radial velocity observations from the 22 May 2011 Ada, Oklahoma, tornadic supercell storm are assimilated over a 45-min period using observations from the experimental PAR located in Norman, Oklahoma, and the operational WSR-88D radar at Oklahoma City, Oklahoma. The radar observations are assimilated into the ARPS model within a heterogeneous mesoscale environment and 1-h ensemble forecasts are generated from analyses every 15 min. With a 30-min assimilation period, the PAR experiment is able to analyze more realistic storm structures, resulting in higher skill scores and higher probabilities of low-level vorticity that align better with the locations of radar-derived rotation compared with the WSR-88D experiment. Assimilation of PAR observations for a longer 45-min time period generates similar forecasts compared to assimilating WSR-88D observations, indicating that the advantage of rapid-scan PAR is more noticeable over a shorter 30-min assimilation period. An additional experiment reveals that the improved accuracy from the PAR experiment over a shorter assimilation period is mainly due to its high-temporal-frequency sampling capability. These results highlight the benefit of PAR’s rapid-scan capability in storm-scale modeling that can potentially extend severe weather warning lead times.

Abstract

With a variety of programming languages and data formats available, widespread adoption of computing standards by the atmospheric science community is often difficult to achieve. The Sounding and Hodograph Analysis and Research Program in Python (SHARPpy) is an open-source, cross-platform, upper-air sounding analysis and visualization package. SHARPpy is based on the National Oceanic and Atmospheric Administration/Storm Prediction Center’s (NOAA/SPC) in-house analysis package, SHARP, and is the result of a collaborative effort between forecasters at the SPC and students at the University of Oklahoma’s School of Meteorology. The major aim of SHARPpy is to provide a consistent framework for sounding analysis that is available to all. Nearly all routines are written to be as consistent as possible with the methods researched, tested, and developed in the SPC, which sets this package apart from other sounding analysis tools.

SHARPpy was initially demonstrated and released to the atmospheric community at the American Meteorological Society (AMS) Annual Meeting in 2012, and an updated and greatly expanded version was released at the AMS Annual Meeting in 2015. Since this release, SHARPpy has been adopted by a variety of operational and research meteorologists across the world. In addition, SHARPpy’s open-source nature enables collaborations between other developers, resulting in major additions to the program.

Abstract

During the winter of 2020/21 an ensemble of FV3-LAM forecasts was produced over the contiguous United States for the Winter Weather Experiment using five physics suites. These forecasts are evaluated with the goal of optimizing physics parameterizations within the future operational Rapid Refresh Forecast System (RRFS) in the Unified Forecast System (UFS) realm and for selecting suitable physics suites for a multiphysics RRFS ensemble. The five physics suites have different combinations of land surface models (LSMs), planetary boundary layer (PBL) parameterizations, and surface layer schemes, chosen from those used in current and possible future operational systems and likely to be supported in the operational UFS. Full-season evaluation reveals a persistent near-surface cold bias in the U.S. Northeast from one suite and a nighttime warm bias in the southern Great Plains in another suite, while other suites have smaller biases. A representative case is chosen to diagnose the cause for each of these biases using sensitivity simulations with different physics combinations or modified parameters and verified with additional mesonet observations. The cold bias in the Northeast is attributed to aspects of the Noah-MP LSM over snow cover, where Noah-MP simulates lower soil water content, and thus lower thermal conductivity than other LSMs, leading to less upward ground heat flux during nighttime and consequently lower surface temperature. The nighttime warm bias found in the southern Great Plains is attributed to overestimation of vertical mixing in the K-profile-based eddy-diffusivity mass-flux (K-EDMF) PBL scheme and insufficient land–atmospheric coupling from the GFS surface layer scheme over short vegetation. A few key parameters driving these systematic biases are identified.

Abstract

To help inform physics configuration decisions and help design and optimize a multi-physics Rapid Refresh Forecasting System (RRFS) ensemble to be used operationally by the National Weather Service, five FV3-LAM-based convection allowing forecasts were run on 35 cases between October 2020 and March 2021. These forecasts used ∼3-km grid spacing on a CONUS domain with physics configurations including Thompson, NSSL, and Ferrier–Aligo microphysics schemes, Noah, RUC, and NoahMP land surface models, and MYNN-EDMF, K-EDMF, and TKE-EDMF PBL schemes. All forecasts were initialized from the 0000 UTC GFS analysis and run for 84 h. Also, a subset of 8 cases were run with 15 combinations of physics options, also including the Morrison–Gettelman microphysics and Shin–Hong PBL schemes, to help attribute behaviors to individual schemes and isolate the main contributors of forecast errors. Evaluations of both sets of forecasts find that the CONUS-wide 24-h precipitation > 1 mm is positively biased across all five forecasts. NSSL microphysics displays a low bias in QPF along the Gulf Coast. Analyses show that it produces smaller raindrops prone to evaporation. Additionally, TKE-EDMF PBL in combination with Thompson microphysics displays a positive bias in precipitation over the Great Lakes and in the ocean near Florida due to higher latent heat fluxes calculated over water. Furthermore, the K-EDMF PBL scheme produces temperature errors that result in a negative bias in snowfall over the southern Mountain West. Finally, recommendations for which physics schemes to use in future suites and the RRFS ensemble are discussed.

Abstract

During the summers of 2016 and 2017, the Center for Analysis and Prediction of Storms (CAPS) ran real-time storm-scale ensemble forecasts (SSEFs) in support of the Hydrometeorology Testbed (HMT) Flash Flood and Intense Rainfall (FFaIR) experiment. These forecasts, using WRF-ARW and Nonhydrostatic Mesoscale Model on the B-grid (NMMB) in 2016, and WRF-ARW and GFDL Finite Volume Cubed-Sphere Dynamical Core (FV3) in 2017, covered the contiguous United States at 3-km horizontal grid spacing, and supported the generation and evaluation of precipitation forecast products, including ensemble probabilistic products. Forecasts of 3-h precipitation accumulation are evaluated. Overall, the SSEF produces skillful 3-h accumulated precipitation forecasts, with ARW members generally outperforming NMMB members and the single FV3 member run in 2017 outperforming ARW members; these differences are significant at some forecast hours. Statistically significant differences exist in the performance, in terms of bias and ETS, among subensembles of members sharing common microphysics and PBL schemes. Year-to-year consistency is higher for PBL subensembles than for microphysical subensembles. Probability-matched (PM) ensemble mean forecasts outperform individual members, while the simple ensemble mean exhibits substantial bias. A newly developed localized probability-matched (LPM) ensemble mean product was produced in 2017; compared to the simple ensemble mean and the conventional PM mean, the LPM mean exhibits improved retention of small-scale structures, evident in both 2D forecast fields and variance spectra. Probabilistic forecasts of precipitation exceeding flash flood guidance (FFG) or thresholds associated with recurrence intervals (RI) ranging from 10 to 100 years show utility in predicting regions of flooding threat, but generally overpredict the occurrence of such events; however, they may still be useful in subjective flash flood risk assessment.

Abstract

A potential replacement candidate for the aging operational WSR-88D infrastructure currently in place is the phased array radar (PAR) system. The current WSR-88Ds take ~5 min to produce a full volumetric scan of the atmosphere, whereas PAR technology allows for full volumetric scanning of the same atmosphere every ~1 min. How this increase in temporal frequency of radar observations might affect the National Severe Storms Laboratory’s (NSSL) Warn-on-Forecast system (WoFS), which is a storm-scale ensemble data assimilation and forecast system for severe convective weather, is unclear. Since radar data assimilation is critical for the WoFS, this study explores the optimal temporal frequency of PAR observations for storm-scale data assimilation using the 31 May 2013 El Reno, Oklahoma, tornadic supercell event. The National Severe Storms Laboratory’s National Weather Radar Testbed PAR in Norman, Oklahoma, began scanning this event more than an hour before the first (and strongest) tornado developed near El Reno, and scanned most of the tornadic supercell’s evolution. Several experiments using various cycling and data frequencies to synchronously and asynchronously assimilate these PAR observations are conducted to produce analyses and very short-term forecasts of the El Reno supercell. Forecasts of low-level reflectivity and midlevel updraft helicity are subjectively evaluated and objectively verified using spatial and object-based techniques. Results indicate that assimilating more frequent PAR observations can lead to more accurate analyses and probabilistic forecasts of the El Reno supercell at longer lead times. Hence, PAR is a promising radar platform for WoFS.

Abstract

A potential replacement candidate for the aging operational WSR-88D infrastructure currently in place is the phased array radar (PAR) system. The current WSR-88Ds take ~5 min to produce a full volumetric scan of the atmosphere, whereas PAR technology allows for full volumetric scanning of the same atmosphere every ~1 min. How this increase in temporal frequency of radar observations might affect the National Severe Storms Laboratory’s (NSSL) Warn-on-Forecast system (WoFS), which is a storm-scale ensemble data assimilation and forecast system for severe convective weather, is unclear. Since radar data assimilation is critical for the WoFS, this study explores the optimal temporal frequency of PAR observations for storm-scale data assimilation using the 31 May 2013 El Reno, Oklahoma, tornadic supercell event. The National Severe Storms Laboratory’s National Weather Radar Testbed PAR in Norman, Oklahoma, began scanning this event more than an hour before the first (and strongest) tornado developed near El Reno, and scanned most of the tornadic supercell’s evolution. Several experiments using various cycling and data frequencies to synchronously and asynchronously assimilate these PAR observations are conducted to produce analyses and very short-term forecasts of the El Reno supercell. Forecasts of low-level reflectivity and midlevel updraft helicity are subjectively evaluated and objectively verified using spatial and object-based techniques. Results indicate that assimilating more frequent PAR observations can lead to more accurate analyses and probabilistic forecasts of the El Reno supercell at longer lead times. Hence, PAR is a promising radar platform for WoFS.

Abstract

Several data assimilation and forecast experiments are undertaken to determine the impact of special observations taken during the second Verification of the Origins of Rotation in Tornadoes Experiment (VORTEX2) on forecasts of the 5 June 2009 Goshen County, Wyoming, supercell. The data used in these experiments are those from the Mobile Weather Radar, 2005 X-band, Phased Array (MWR-05XP); two mobile mesonets (MM); and several mobile sounding units. Data sources are divided into “routine,” including those from operational Weather Surveillance Radar-1988 Dopplers (WSR-88Ds) and the Automated Surface Observing System (ASOS) network, and “special” observations from the VORTEX2 project.

VORTEX2 data sources are denied individually from a total of six ensemble square root filter (EnSRF) data assimilation and forecasting experiments. The EnSRF data assimilation uses 40 ensemble members on a 1-km grid nested inside a 3-km grid. Each experiment assimilates data every 5 min for 1 h, followed by a 1-h forecast. All experiments are able to reproduce the basic evolution of the supercell, though the impact of the VORTEX2 observations was mixed. The VORTEX2 sounding data decreased the mesocyclone intensity in the latter stages of the forecast, consistent with observations. The MWR-05XP data increased the forecast vorticity above approximately 1 km AGL in all experiments and had little impact on forecast vorticity below 1 km AGL. The MM data had negative impacts on the intensity of the low-level mesocyclone, by decreasing the vertical vorticity and indirectly by decreasing the buoyancy of the inflow.

Abstract

Assimilation of dual-polarization (dual-pol) observations provides more accurate storm-scale analyses to initialize forecasts of severe convective thunderstorms. This study investigates the impact assimilating experimental sector-scan dual-pol observations has on storm-scale ensemble forecasts and how this impact changes over different data assimilation (DA) windows using the ensemble Kalman filter (EnKF). Ensemble forecasts are initialized after 30, 45, and 60 min of DA for two sets of experiments that assimilate either reflectivity and radial velocity only (EXPZ) or reflectivity and radial velocity plus differential reflectivity (EXPZZDR). This study uses the 31 May 2013 Oklahoma event, which included multiple storms that produced tornadoes and severe hail, with a focus on two storms that impacted El Reno and Stillwater during the event. The earliest initialized forecast of EXPZZDR better predicts the evolution of the El Reno storm than EXPZ, but the two sets of experiments become similar at subsequent forecast times. However, the later EXPZZDR forecasts of the Stillwater storm, which organized toward the end of the DA window, produce improved results compared to EXPZ, in which the storm is less intense and weakens. Evaluation of forecast products for supercell mesocyclones [updraft helicity (UH)] and hail show similar results, with earlier EXPZZDR forecasts better predicting the UH swaths of the El Reno storm and later forecasts producing improved UH and hail swaths for the Stillwater storm. The results indicate that the assimilation of Z _DR over fewer DA cycles can produce improved forecasts when DA windows sufficiently cover storms during their initial development and organization.