Extracting Unique Information from High-Resolution Forecast Models: Monitoring Selected Fields and Phenomena Every Time Step

John S. Kain NOAA/National Severe Storms Laboratory, Norman, Oklahoma

Search for other papers by John S. Kain in
Current site
Google Scholar
PubMed
Close
,
Scott R. Dembek Cooperative Institute for Mesoscale Meteorological Studies, University of Oklahoma, and NOAA/OAR/National Severe Storms Laboratory, Norman, Oklahoma
Universities Space Research Association/Short-term Prediction Research and Transition Center, Huntsville, Alabama

Search for other papers by Scott R. Dembek in
Current site
Google Scholar
PubMed
Close
,
Steven J. Weiss NOAA/NWS/Storm Prediction Center, Norman, Oklahoma

Search for other papers by Steven J. Weiss in
Current site
Google Scholar
PubMed
Close
,
Jonathan L. Case ENSCO Inc./SPoRT Center, Huntsville, Alabama

Search for other papers by Jonathan L. Case in
Current site
Google Scholar
PubMed
Close
,
Jason J. Levit * NOAA/NWS/Aviation Weather Center, Kansas City, Missouri

Search for other papers by Jason J. Levit in
Current site
Google Scholar
PubMed
Close
, and
Ryan A. Sobash School of Meteorology, University of Oklahoma, Norman, Oklahoma

Search for other papers by Ryan A. Sobash in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

A new strategy for generating and presenting model diagnostic fields from convection-allowing forecast models is introduced. The fields are produced by computing temporal-maximum values for selected diagnostics at each horizontal grid point between scheduled output times. The two-dimensional arrays containing these maximum values are saved at the scheduled output times. The additional fields have minimal impacts on the size of the output files and the computation of most diagnostic quantities can be done very efficiently during integration of the Weather Research and Forecasting Model. Results show that these unique output fields facilitate the examination of features associated with convective storms, which can change dramatically within typical output intervals of 1–3 h.

Corresponding author address: John S. Kain, National Severe Storms Laboratory, 120 David L. Boren Blvd., Norman, OK 73072. Email: jack.kain@noaa.gov

Abstract

A new strategy for generating and presenting model diagnostic fields from convection-allowing forecast models is introduced. The fields are produced by computing temporal-maximum values for selected diagnostics at each horizontal grid point between scheduled output times. The two-dimensional arrays containing these maximum values are saved at the scheduled output times. The additional fields have minimal impacts on the size of the output files and the computation of most diagnostic quantities can be done very efficiently during integration of the Weather Research and Forecasting Model. Results show that these unique output fields facilitate the examination of features associated with convective storms, which can change dramatically within typical output intervals of 1–3 h.

Corresponding author address: John S. Kain, National Severe Storms Laboratory, 120 David L. Boren Blvd., Norman, OK 73072. Email: jack.kain@noaa.gov

1. Introduction

Traditionally, output from numerical weather prediction (NWP) models has been presented to forecasters as a series of snapshots in time (an exception is accumulated precipitation). As model resolution and transmission bandwidth have increased, the time interval between these snapshots has decreased. For example, output from the early operational NWP models in the United States, such as the Limited-area Fine Mesh (LFM) model was broadly available only at 12-hourly intervals, but some output from current models is supplied hourly to forecasters at the National Weather Service (NWS) and elsewhere.

For most forecasting applications, hourly output is adequate because the evolution of common larger-scale features of interest (i.e., fronts, jet streaks, low and high pressure centers, etc.) is well sampled by the hourly frequency. Furthermore, this frequency is a pragmatic choice because the sheer volume of data associated with more frequent output files would exceed the capacity of current dissemination, processing, and storage systems. However, as NWP applications move to higher resolution, the features of interest begin to change and hourly sampling can become inadequate.

Of particular interest here are processes associated with deep moist convection. Until recently, deep convection has been parameterized in operational NWP models, but in the last several years relatively high-resolution convection-allowing models (CAMs) have emerged as valuable forecast tools (e.g., Weiss et al. 2008; Dixon et al. 2009). The term “convection allowing” is used to emphasize that these models are not truly convection resolving—that would require grid spacing on the order of O(100m) or less (Bryan et al. 2003; Craig and Dörnbrack 2008)—but CAMs can still represent convective overturning explicitly and are capable of providing valuable clues about convective storm activity. For example, CAMs with grid spacing as coarse as 4 km can discriminate between the different mesoscale organizational modes of convection (e.g., Done et al. 2004; Weisman et al. 2008) and they can even provide useful predictions of supercells (Kain et al. 2008), which are associated with a wide variety of severe weather (i.e., tornadoes, large hail, and damaging winds).

Simulated features such as these often develop, move, and vary in intensity on time scales commonly measured in minutes, not hours. Thus, there is strong motivation to monitor their behavior at a higher frequency than hourly output provides. This could be done by increasing the output frequency of all or selected fields, but a somewhat different approach is taken here. Specifically, this study describes a strategy for monitoring small-scale, rapidly changing features every model time step (24 s in this application) between regular output times. The individual grid-point temporal-maximum values from each hour are stored in two-dimensional fields that are saved at the regular hourly output intervals, providing a useful perspective on the maximum intensity and track of strong phenomena. The focus is on severe convection, but the method could easily be applied to study and/or predict other phenomena as well. The method is described in the next section. The third section contains sample applications and describes the potential value added by the unique output fields. Results are summarized in section 4.

2. Methodology

This study was conducted using the Weather Research and Forecasting Model (WRF; Skamarock et al. 2005) as part of a collaborative effort between the National Severe Storms Laboratory (NSSL) and the Storm Prediction Center (SPC), hence its focus on severe convective weather. NSSL has been running the WRF (hereafter WRF-NSSL) daily in real time since late 2006 (Table 1). Hourly output is posted on the NSSL Web site (http://www.nssl.noaa.gov/wrf/) and transmitted directly to the SPC and several NWS Weather Forecast Offices. Model forecasts are made daily out to 36 h, using 0000 UTC initial and lateral boundary conditions from the operational North American Mesoscale (NAM) model (Rogers et al. 2009), over a nearly conterminous United States (CONUS) scale domain.

Currently, a total of eight diagnostic quantities are monitored at every time step during model integration and temporal-maximum values of each are computed at every point on the model’s horizontal grid. At prescribed hourly output times, the eight two-dimensional arrays containing the maximum values (hereafter hourly maximum fields, HMFs) are appended to the full output file and the arrays are reset to zero. The diagnostic quantities include the following:

  • updraft and downdraft velocities below the 400-hPa level (UP and DN, respectively; units of m s−1), which have obvious implications regarding the intensity of convective overturning;

  • the simulated reflectivity computed at the lowest model level (RF; units of dBZ), as with updraft and downdraft velocities, this field is expected to be related to the intensity of convection;

  • wind speed at 10 m above ground level (AGL) (UU; units of m s−1), which may be helpful in predicting the magnitude of convectively generated surface wind gusts;

  • updraft helicity between 2 and 5 km AGL (UH; units of m2 s−2), which is designed to detect mesocyclones in model storms as they may be useful surrogates for supercells (Kain et al. 2008);

  • vertically integrated graupel (GR; units of kg m−2, or depth of equivalent liquid in mm), which may be a useful predictor of hail; and

  • total lightning threats 1 and 2 (L1 and L2), based on upward graupel flux at −15°C and vertically integrated total ice, respectively (McCaul et al. 2009).

Initial subjective impressions suggest that these quantities may have some utility, as forecast guidance and quantitative verification of each has begun. For the purpose of demonstration in this paper, the UH and UU fields are spotlighted since they have received the most scrutiny in the operational forecasting environment. The GR field, which has been available for a shorter time period, is also highlighted for one case. Conceptually at least, these three fields from a 4-km CAM may be the closest analogs to the severe phenomena that SPC forecasters must predict: tornadoes, damaging winds, and large hail.

3. Results

Three recent severe weather events were selected to demonstrate the potential utility of the HMFs. In each case severe convective activity occurred primarily after 1800 UTC, 18 h into the WRF forecast. The output shown here was available in real time on the day of each event.

a. 3 June 2008

On this day there were multiple severe weather events across the eastern half of the United States (Fig. 1), but for the purposes of this study the focus is on the activity along the Oklahoma–Kansas border. In its early stages, this activity consisted of a series of isolated supercellular thunderstorms in the Oklahoma panhandle. Within the first few hours, one of these cells emerged as a dominant, large-hail-producing supercell, then the cluster consolidated, appeared to grow upscale dynamically, and eventually developed into a mesoscale convective system (MCS) that produced a swath of damaging winds with reported speeds of up to 65 kt (33 m s−1; see Figs. 1 and 2a–c). In a broad sense, the WRF-NSSL initialization from the night before performed exceptionally well in capturing the evolution of this convective system (cf. Figs. 2a–c and 2d–f). Furthermore, the hourly maximum UH and UU fields provided additional information that could help forecasters gain more insight into system development, including the convective-mode transition from supercells to MCSs. For example, the UH tracks in Fig. 2g indicate that the dominant storm in the eastern Oklahoma panhandle at 2300 UTC was long lived and strongly rotating with UH values exceeding 190 m2 s−2. [This value is quite high; in experimental guidance products used by forecasters at the SPC, grid points are highlighted if UH exceeds 50 m2 s−2 with the model configuration used here. See Kain et al. (2008) for more information regarding the sensitivity of UH to model configuration.] By 0100 UTC UH values had weakened and two parallel track segments were evident (Fig. 2h), suggesting that storm splitting had occurred over the previous hour. By 0300 UTC, individual mesocyclones were much weaker and shorter lived (Fig. 2i). Meanwhile, the UU field suggested that during the early stages the strongest winds were associated with individual storms and were relatively limited in their areal coverage (Figs. 2j and 2k). By 0300 UTC (Fig. 2l), however, the simulated wind field contained a broad swath of very strong winds, exceeding 35 m s−1 in some areas, consistent with the upscale growth and rapidly spreading cold pool (not shown) that was evident in both the observations and model output.

b. 5 February 2008

On this day there was a major severe weather outbreak across the mid-South and Tennessee Valley (Figs. 3a and 3b), including 10 deadly tornadoes resulting in a total of 57 fatalities. When compared with observations on an hour-by-hour basis, the WRF-NSSL forecast had many flaws on this day. For example, it contained errors in the placement and timing of several important convective features (not shown). Yet, in a more general sense, the forecast was quite good at highlighting the corridor where the threat for significant severe weather was the greatest, as well as the potential intensity of individual storms. This is illustrated by showing the maximum UH at each grid point during the entire 24-h period—essentially combining hourly segments of maximum UH. The image produced by this process shows that, over the course of this event, the model predicted a swath of long-lived mesocyclones over much the same region where long-track supercells and significant killer tornadoes were observed (Fig. 3c). This example demonstrates that convective-mode and intensity forecasts from CAMs can be very useful, even on some occasions when there are significant errors in predicting the specific timing and location of an individual storm.

c. 23 April 2009

On this day the dominant severe weather event in the United States was a concentrated hail episode in northern Georgia (Fig. 4a). Hailstones ranged up to golf-ball size and they completely covered the ground in some places. Simulated reflectivity fields from the WRF-NSSL predicted the timing and location of storms fairly well, perhaps an hour or two early on the timing and with a slight displacement to the southwest (not shown). Despite these differences, the HMFs once again appeared to provide useful guidance regarding the dominant severe weather threat associated with these storms. For example, the maximum UH values over the course of the event were relatively low (<55 m2 s−2) (Fig. 4b), suggesting that midlevel rotation would be comparatively weak in any storms that formed. Likewise, maximum 10-m wind speeds (UU) were moderately strong in some locations, but less than 22.5 m s−1 throughout the event (Fig. 4c). However, the GR diagnostic did yield intriguing results on this day (Fig. 4d). This field was not available during 2008 and meaningful threshold values are less clearly defined than those for UH and UU, but empirical evidence derived from monitoring daily output during 2009 suggests that the maximum values of 35–45 kg m−2 shown in Fig. 4d are comparable to those predicted for other large-hail events. Additional analysis of the GR field is needed for confirmation, but a combined assessment of the UH, UU, and GR fields provided at least a reasonable indication that large hail would be the predominant threat on this day.

4. Summary

A new strategy for extracting, processing, and collecting output from CAMs is presented. It is designed to provide important clues about the formation, evolution, and intensity of convective storm features and related circulations in the interval between model output times, while having minimal impacts on the size of the model output files and the computational demands of the model. Our method involves monitoring specific diagnostic quantities in every grid column during every model time step, but saving only two-dimensional arrays containing grid-point temporal-maximum values of each quantity. The two-dimensional fields containing the maximum values are appended to regularly scheduled output files. These fields have been produced since February 2008 during daily runs of the WRF-NSSL, a CONUS-scale real-time modeling system with 4-km horizontal grid spacing. Since the current output interval for the WRF-NSSL is 1 h, these fields are called hourly maximum fields, or HMFs, in this paper.

For this study, three severe weather events were selected to demonstrate the potential utility of the HMF concept. In these events the unique fields appear to provide value-added guidance regarding such factors as to whether long-lived supercells might develop or whether large hail from mainly nonsupercellular storms might be the predominant severe weather threat. Skillful guidance in these areas could prove to be very valuable for severe weather forecasters.

At this time, the potential value of these HMFs is based only on subjective assessments. Clearly, a thorough quantitative evaluation and calibration, based on the correspondence of each HMF to observed severe weather, will be needed before the full potential of these unique output fields can be realized. Such an evaluation is under way (Sobash et al. 2009), but the HMFs are already being used by forecasters at the SPC and elsewhere because they have proven to be useful in operations on numerous occasions. Based on these initial promising results, the HMF concept has been adopted by a number of our partners, including the National Centers for Environmental Prediction/Environmental Modeling Center (NCEP/EMC; M. Pyle 2010, personal communication), the National Oceanic and Atmospheric Administration/Earth Systems Research Laboratory/Global Sciences Division (NOAA/ESRL/GSD; S. Weygandt 2009, personal communication), the Air Force Weather Agency (AFWA; E. Kuchera 2010, personal communication), the University of Oklahoma’s Center for Analysis and Prediction of Storms (OU–CAPS, F. Kong 2010, personal communication), the National Center for Atmospheric Research (NCAR; M. Weisman 2010, personal communication), and has also been incorporated into the National Weather Service WRF Environmental Modeling System framework (Rozumalski 2010).

This effort represents an important first step in exploring ways of extracting new output fields and/or computing new diagnostics from CAMs. This is significant because the emergence of high-resolution models means an entirely new subset of phenomena can now be depicted in forecast models. A major challenge is to identify and extract the information that can provide the most useful guidance to forecasters.

Acknowledgments

We thank Jimy Dudhia of NCAR for his help in modifying WRF model code to extract the HMFs. We are grateful to Phillip Bothwell and Chris Melick of the SPC for retrieving radar data for selected events. We are grateful to two anonymous reviewers and Mike Coniglio of NSSL for helpful comments on an earlier version of this manuscript. Partial funding was provided by NOAA/Office of Oceanic and Atmospheric Research under NOAA–University of Oklahoma Cooperative Agreement NA17RJ1227, U.S. Department of Commerce.

REFERENCES

  • Bryan, G. H., Wyngaard J. C. , and Fritsch J. M. , 2003: Resolution requirements for the simulation of deep moist convection. Mon. Wea. Rev., 131 , 23942416.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Craig, G. C., and Dörnbrack A. , 2008: Entrainment in cumulus clouds: What resolution is cloud resolving? J. Atmos. Sci., 65 , 39783988.

  • Dixon, M., Li Z. , Lean H. , Roberts N. , and Ballard S. , 2009: Impact of data assimilation on forecasting convection over the United Kingdom using a high-resolution version of the Met Office Unified Model. Mon. Wea. Rev., 137 , 15621584.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Done, J., Davis C. , and Weisman M. , 2004: The next generation of NWP: Explicit forecasts of convection using the Weather Research and Forecasting (WRF) model. Atmos. Sci. Lett., 5 (6) 110117.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kain, J. S., and Coauthors, 2008: Some practical considerations regarding horizontal resolution in the first generation of operational convection-allowing NWP. Wea. Forecasting, 23 , 931952.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • McCaul, E. W., Goodman S. J. , LaCasse K. M. , and Cecil D. J. , 2009: Forecasting lightning threat using cloud-resolving model simulations. Wea. Forecasting, 24 , 709729.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rogers, E., and Coauthors, 2009: The NCEP North American Mesoscale modeling system: Recent changes and future plans. Preprints, 23rd Conf. on Weather Analysis and Forecasting/19th Conf. on Numerical Weather Prediction, Omaha, NE, Amer. Meteor. Soc., 2A.4. [Available online at http://ams.confex.com/ams/pdfpapers/154114.pdf].

    • Search Google Scholar
    • Export Citation
  • Rozumalski, R. A., cited. 2010: A nearly complete guide to the WRF EMS version 3.1. NOAA/National Weather Service/Forecast Decision Training Branch. [Available online at http://strc.comet.ucar.edu/wrf/usersguide/wrfems_userguide.htm].

    • Search Google Scholar
    • Export Citation
  • Skamarock, W. C., Klemp J. B. , Dudhia J. , Gill D. O. , Barker D. M. , Wang W. , and Powers J. G. , 2005: A description of the Advanced Research WRF version 2. NCAR Tech. Note NCAR/TN-468 STR, 88 pp. [Available from UCAR Communications, P.O. Box 3000, Boulder, CO 80307].

    • Search Google Scholar
    • Export Citation
  • Sobash, R. A., Kain J. S. , Bright D. R. , Dean A. R. , Coniglio M. C. , Weiss S. J. , and Levit J. J. , 2009: Forecast guidance for severe thunderstorms based on identification of extreme phenomena in convection-allowing model forecasts. Preprints, 23rd Conf. on Weather Analysis and Forecasting/19th Conf. on Numerical Weather Prediction, Omaha, NE, Amer. Meteor. Soc., 4B.6. [Available online at http://ams.confex.com/ams/pdfpapers/154328.pdf].

    • Search Google Scholar
    • Export Citation
  • Weisman, M. L., Davis C. , Wang W. , Manning K. W. , and Klemp J. B. , 2008: Experiences with 0–36-h explicit convective forecasts with the WRF-ARW model. Wea. Forecasting, 23 , 407437.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Weiss, S. J., Pyle M. E. , Janjic Z. , Bright D. R. , and DiMego G. J. , 2008: The operational high-resolution window WRF model runs at NCEP: Advantages of multiple model runs for severe convective weather forecasting. Preprints, 24th Conf. on Severe Local Storms, Savannah, GA, Amer. Meteor. Soc., P10.8. [Available online at http://ams.confex.com/ams/pdfpapers/142192.pdf].

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Locations of severe storm reports for the period 1200–1159 UTC 3–4 Jun 2008 from the preliminary assessment compiled by the SPC. Blue dots indicate convective wind damage and/or wind gusts ≥50 kt (26 m s−1), green dots indicate hail ≥0.75 in. (19 mm) in diameter, and red dots indicate tornadoes. Large black squares denote wind gusts ≥65 kt (33 m s−1) while large black triangles indicate hail ≥2 in. (51 mm) in diameter.

Citation: Weather and Forecasting 25, 5; 10.1175/2010WAF2222430.1

Fig. 2.
Fig. 2.

Observed and simulated evolution patterns of a convective complex from 2300 UTC 3 Jun 2008 through 0300 UTC 4 Jun 2008, showing observed lowest-elevation-angle reflectivity (top row, dBZ), simulated 1-km AGL reflectivity (second row, dBZ), and simulated hourly-maximum values of UH (third row, m2 s−2), and 10-m wind speed (bottom row, m s−1) from the WRF-NSSL. Simulated 10-m AGL wind vectors are also shown in the bottom three rows.

Citation: Weather and Forecasting 25, 5; 10.1175/2010WAF2222430.1

Fig. 3.
Fig. 3.

(a) As in Fig. 1, but for 1200–1159 UTC 5–6 Feb 2008. Also shown are (b) tornado tracks and enhanced Fujita (EF) scale intensities for this same time period based on NWS damage surveys and (c) temporal-maximum grid-point UH values (m2 s−2) from the WRF-NSSL during the same time period.

Citation: Weather and Forecasting 25, 5; 10.1175/2010WAF2222430.1

Fig. 4.
Fig. 4.

(a) As in Fig. 1, but for 1200–1159 UTC 23–24 Apr 2009. (b)–(d) The temporal-maximum grid-point values of UH (m2 s−2), 10-m wind speed (m s−1), and vertically integrated graupel (kg m−2), respectively, from the WRF-NSSL during the same time period.

Citation: Weather and Forecasting 25, 5; 10.1175/2010WAF2222430.1

Table 1.

Model configuration for the daily WRF-NSSL forecasts. NAM is the operational North American Mesoscale model run by NCEP/EMC. Descriptions of the different WRF physical parameterizations can be found online (http://www.mmm.ucar.edu/wrf/users/docs/user_guide_V3/users_guide_chap5.htm#Phys).

Table 1.
Save
  • Bryan, G. H., Wyngaard J. C. , and Fritsch J. M. , 2003: Resolution requirements for the simulation of deep moist convection. Mon. Wea. Rev., 131 , 23942416.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Craig, G. C., and Dörnbrack A. , 2008: Entrainment in cumulus clouds: What resolution is cloud resolving? J. Atmos. Sci., 65 , 39783988.

  • Dixon, M., Li Z. , Lean H. , Roberts N. , and Ballard S. , 2009: Impact of data assimilation on forecasting convection over the United Kingdom using a high-resolution version of the Met Office Unified Model. Mon. Wea. Rev., 137 , 15621584.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Done, J., Davis C. , and Weisman M. , 2004: The next generation of NWP: Explicit forecasts of convection using the Weather Research and Forecasting (WRF) model. Atmos. Sci. Lett., 5 (6) 110117.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kain, J. S., and Coauthors, 2008: Some practical considerations regarding horizontal resolution in the first generation of operational convection-allowing NWP. Wea. Forecasting, 23 , 931952.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • McCaul, E. W., Goodman S. J. , LaCasse K. M. , and Cecil D. J. , 2009: Forecasting lightning threat using cloud-resolving model simulations. Wea. Forecasting, 24 , 709729.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Rogers, E., and Coauthors, 2009: The NCEP North American Mesoscale modeling system: Recent changes and future plans. Preprints, 23rd Conf. on Weather Analysis and Forecasting/19th Conf. on Numerical Weather Prediction, Omaha, NE, Amer. Meteor. Soc., 2A.4. [Available online at http://ams.confex.com/ams/pdfpapers/154114.pdf].

    • Search Google Scholar
    • Export Citation
  • Rozumalski, R. A., cited. 2010: A nearly complete guide to the WRF EMS version 3.1. NOAA/National Weather Service/Forecast Decision Training Branch. [Available online at http://strc.comet.ucar.edu/wrf/usersguide/wrfems_userguide.htm].

    • Search Google Scholar
    • Export Citation
  • Skamarock, W. C., Klemp J. B. , Dudhia J. , Gill D. O. , Barker D. M. , Wang W. , and Powers J. G. , 2005: A description of the Advanced Research WRF version 2. NCAR Tech. Note NCAR/TN-468 STR, 88 pp. [Available from UCAR Communications, P.O. Box 3000, Boulder, CO 80307].

    • Search Google Scholar
    • Export Citation
  • Sobash, R. A., Kain J. S. , Bright D. R. , Dean A. R. , Coniglio M. C. , Weiss S. J. , and Levit J. J. , 2009: Forecast guidance for severe thunderstorms based on identification of extreme phenomena in convection-allowing model forecasts. Preprints, 23rd Conf. on Weather Analysis and Forecasting/19th Conf. on Numerical Weather Prediction, Omaha, NE, Amer. Meteor. Soc., 4B.6. [Available online at http://ams.confex.com/ams/pdfpapers/154328.pdf].

    • Search Google Scholar
    • Export Citation
  • Weisman, M. L., Davis C. , Wang W. , Manning K. W. , and Klemp J. B. , 2008: Experiences with 0–36-h explicit convective forecasts with the WRF-ARW model. Wea. Forecasting, 23 , 407437.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Weiss, S. J., Pyle M. E. , Janjic Z. , Bright D. R. , and DiMego G. J. , 2008: The operational high-resolution window WRF model runs at NCEP: Advantages of multiple model runs for severe convective weather forecasting. Preprints, 24th Conf. on Severe Local Storms, Savannah, GA, Amer. Meteor. Soc., P10.8. [Available online at http://ams.confex.com/ams/pdfpapers/142192.pdf].

    • Search Google Scholar
    • Export Citation
  • Fig. 1.

    Locations of severe storm reports for the period 1200–1159 UTC 3–4 Jun 2008 from the preliminary assessment compiled by the SPC. Blue dots indicate convective wind damage and/or wind gusts ≥50 kt (26 m s−1), green dots indicate hail ≥0.75 in. (19 mm) in diameter, and red dots indicate tornadoes. Large black squares denote wind gusts ≥65 kt (33 m s−1) while large black triangles indicate hail ≥2 in. (51 mm) in diameter.

  • Fig. 2.

    Observed and simulated evolution patterns of a convective complex from 2300 UTC 3 Jun 2008 through 0300 UTC 4 Jun 2008, showing observed lowest-elevation-angle reflectivity (top row, dBZ), simulated 1-km AGL reflectivity (second row, dBZ), and simulated hourly-maximum values of UH (third row, m2 s−2), and 10-m wind speed (bottom row, m s−1) from the WRF-NSSL. Simulated 10-m AGL wind vectors are also shown in the bottom three rows.

  • Fig. 3.

    (a) As in Fig. 1, but for 1200–1159 UTC 5–6 Feb 2008. Also shown are (b) tornado tracks and enhanced Fujita (EF) scale intensities for this same time period based on NWS damage surveys and (c) temporal-maximum grid-point UH values (m2 s−2) from the WRF-NSSL during the same time period.

  • Fig. 4.

    (a) As in Fig. 1, but for 1200–1159 UTC 23–24 Apr 2009. (b)–(d) The temporal-maximum grid-point values of UH (m2 s−2), 10-m wind speed (m s−1), and vertically integrated graupel (kg m−2), respectively, from the WRF-NSSL during the same time period.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 525 116 14
PDF Downloads 374 84 14