• Crum, T. D., and R. L. Alberty, 1993: The WSR-88D and the WSR-88D operational support facility. Bull. Amer. Meteor. Soc.,74, 1669–1687.

    • Crossref
    • Export Citation
  • Doswell, C. A., III, 1985: The operational meteorology of convective weather. Vol. II: Storm scale analysis. NOAA Tech. Memo. ERL ESG-15, Boulder, CO, 102–103.

  • Eilts, M. D., and Coauthors, 1996: Severe weather warning decision support system. Preprints, 18th Conf. on Severe Local Storms, San Francisco, CA, Amer. Meteor. Soc., 536–540.

  • Howard, K. W., C. L. Dempsey, and D. M. McCollum, 1993: SW Area Monsoon Project: Daily operations summary. NSSL, 435 pp. [Available from National Severe Storms Laboratory, Norman, OK 73069.].

  • Johnson, J. T., and Coauthors 1995: Operational testing of enhanced WSR-88D algorithms and display concepts in National Weather Service offices. Preprints, 27th Conf. on Radar Meteorology, Vail, CO, Amer. Meteor. Soc., 170–172.

  • Maddox, R. A., C. L. Dempsey, and K. W. Howard, 1997: Intense convective storms with little or no lightning over central Arizona—A case of inadvertent weather modification? J. Appl. Meteor.,36, 302–314.

    • Crossref
    • Export Citation
  • NOAA, 1991: Doppler radar meteorological observations, Part C, WSR-88D products and algorithms. Federal Meteorological Handbook, Office of the Federal Coordinator for Meteorological Services and Supporting Research, FCH-H11C-1991, Rockville, MD, 210 pp.

  • Witt, A., and J. T. Johnson, 1993: Operational testing of enhanced WSR-88D algorithms and display concepts in National Weather Service offices. Preprints, 26th Conf. on Radar Meteorology, Norman, OK, Amer. Meteor. Soc., 141–143.

  • View in gallery

    Display from the WDSS system from the KIWA radar (see Eilts et al. 1996). The change with time of various parameters associated with the thunderstorm cell north of Globe, Arizona, is shown in the “trend” windows along the right side of the figure.

  • View in gallery

    A scatterplot of observed storm tops vs radar range for 450 identified storm cells from the KIWA, radar on 20 August 1994.

  • View in gallery

    (a) Radar beam geometry (range vs height) for VCP-11. Beamwidth is 0.95°, and there are 14 different elevation scans with a 5-min update rate, (b) same figure for VCP-21, which has only nine different elevation scans with a 6-min update rate (NOAA 1991).

  • View in gallery

    Uncertainty (m) between the WSR-88D determined height of a reflectivity target and the actual height for (a) VCP-21 and (b) VCP-11. The uncertainty of the height measurement is zero at the beam centerline, and the elevation angles of the various beam centerlines are indicated. Note that the height of features that do not reach the centerline of the 0.5° beam cannot be determined. The height of features located to the left of the centerline of the 19.5° beam cannot be determined.

  • View in gallery

    (Continued)

  • View in gallery

    Vertical reflectivity structure during the life cycle of a pulse-type thunderstorm. Contours are 10-, 30-, and 50-dBZ reflectivity. The solid line denotes the evolution of the height of the top of the 30-dBZ echo core (i.e., WSR-88D storm top). Figure is after Doswell (1985).

  • View in gallery

    Plot of radar-observed storm top and the model storm top for VCP-21 for storm initiation at range 100 km (a) stationary, (b) movement toward the radar at 5 m s−1, and (c) movement toward the radar at 10 m s−1. Data points are at the end of each radar scan and are 6 min apart.

  • View in gallery

    Plot of radar-observed storm top and the model storm top for VCP-21 for storm initiation at range 50 km. Data points are at the end of each radar scan and are 5 min apart. Details are similar to Fig. 6.

  • View in gallery

    Plot of radar-observed storm top and the model storm top for VCP-11 for (a) stationary at range 50 km, (b) movement toward the radar at 5 m s−1 and at 50-km range, and (c) movement toward the radar at 5 m s−1 at range 100 km.

  • View in gallery

    Plot of the radar-observed storm top and the model storm top for (a) VCP-11 with movement toward the radar at 5 m s−1 and storm initiation at range 150 km, and (b) VCP-21 for the same storm.

  • View in gallery

    Plot of the radar-observed storm top and the model storm top for VCP-11 for (a) movement toward the radar at 5 m s−1 and storm initiation at range 25 km and (b) movement toward the radar at 5 m s−1 and storm initiation at range 200 km.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 107 107 4
PDF Downloads 20 20 2

Uncertainties in WSR-88D Measurements and Their Impacts on Monitoring Life Cycles

View More View Less
  • 1 NOAA/Environmental Research Laboratories, National Severe Storms Laboratory, Norman, Oklahoma
  • | 2 Cooperative Institute for Mesoscale Meteorological Studies, Norman, Oklahoma
  • | 3 NOAA/Environmental Research Laboratories, National Severe Storms Laboratory, and Cooperative Institute for Mesoscale Meteorological Studies, Norman, Oklahoma
© Get Permissions
Full access

Abstract

Radar measurement uncertainties associated with storm top, cloud top, and other height measurements are well recognized; however, the authors feel the resulting impacts on the trends of storm features are not as well documented or understood by some users of the WSR-88D system. Detailed examination of radar-measured life cycles of thunderstorms occurring in Arizona indicates substantial limitations in the WSR-88D’s capability to depict certain aspects of storm-height attribute evolution (i.e., life cycle) accurately. These inherent limitations are illustrated using a vertical reflectivity structure model for the life cycle of a simple, “single-pulse” thunderstorm. The life cycle of this simple storm is “scanned” at varying ranges and translation speeds. The results show that radar-determined trends are often substantially different from those of the model storm and that in extreme cases the radar-detected storm and the model storm can have trends in storm-top height of opposite sign. Caution is clearly required by both the operational and research users of some products derived from operational WSR-88D data.

Corresponding author address: Kenneth W. Howard, U.S. Department of Commerce, NOAA/ERL, National Severe Storms Laboratory, 1313 Halley Circle, Norman, OK 73069.

Email: khoward@nssl.uoknor.edu

Abstract

Radar measurement uncertainties associated with storm top, cloud top, and other height measurements are well recognized; however, the authors feel the resulting impacts on the trends of storm features are not as well documented or understood by some users of the WSR-88D system. Detailed examination of radar-measured life cycles of thunderstorms occurring in Arizona indicates substantial limitations in the WSR-88D’s capability to depict certain aspects of storm-height attribute evolution (i.e., life cycle) accurately. These inherent limitations are illustrated using a vertical reflectivity structure model for the life cycle of a simple, “single-pulse” thunderstorm. The life cycle of this simple storm is “scanned” at varying ranges and translation speeds. The results show that radar-determined trends are often substantially different from those of the model storm and that in extreme cases the radar-detected storm and the model storm can have trends in storm-top height of opposite sign. Caution is clearly required by both the operational and research users of some products derived from operational WSR-88D data.

Corresponding author address: Kenneth W. Howard, U.S. Department of Commerce, NOAA/ERL, National Severe Storms Laboratory, 1313 Halley Circle, Norman, OK 73069.

Email: khoward@nssl.uoknor.edu

1. Introduction

The replacement of the U.S. weather radar surveillance system as part of the National Weather Service (NWS) modernization has resulted in unprecedented high-resolution radar observations of convective weather systems for operational use. Prior to the installation of the Weather Surveillance Radar-1988 Doppler (WSR-88D; Crum and Alberty 1993) network, high-resolution digital data depicting the individual life cycles of thunderstorms could only be attained through the use of research radars employed during limited field programs. These data were typically unavailable for operational use.

The operational WSR-88D produces very high rates of digital data and requires computer-based algorithms for the extraction of information critical for monitoring thunderstorms and associated features. A high priority, as the NWS modernization proceeds, is the improvement and enhancement of products and algorithms derived from the NEXRAD raw data stream. It is also important that effective presentation of WSR-88D data and products be available to the operational forecaster.

The National Severe Storms Laboratory (NSSL) has developed, during the past decade, products and algorithms (e.g., storm cell identification and tracking, mesocyclone, and tornado detection algorithms, etc.) in response to the needs of the NWS and the Federal Aviation Administration. The extraction of storm-cell-based attributes, individual storm cell tracks, and the creation of cell-based trends, displayed within an experimental warning decision support system (WDSS; Eilts et al. 1996), provides a potentially powerful tool to monitor the evolution of individual thunderstorms. An example of outputted display from the WDSS system is shown in Fig. 1. The importance and significance of this ability within the WDSS was evident from the initial feedback from NWS personnel during Radar Algorithm and Display System field tests (Witt and Johnson 1993; Johnson et al. 1995). Individual NWS forecasters felt that improvements in warning operations are made possible by using trend information to assess storm severity and life cycle state (e.g., whether the storm is weakening or intensifying). Additionally, application research can be accomplished for large populations of thunderstorm cells using the capabilities and output of WDSS. The careful evaluation of storm attributes, presented in a life cycle format such as with cell-based trends, can help elucidate relationships between storms, their life cycles, and attendant severe weather (e.g., tornadoes, downbursts, lightning frequency, etc.).

Using radar data to monitor thunderstorms in the western United States is both a difficult and new challenge for NWS forecasters. In the West, observations of storms and associated phenomena are often limited by complex terrain, low population densities (i.e., little ground truth information, sparse weather spotter networks), and large expanses (gaps) in radar coverage. Thus, there is little documentation of the occurrence of thunderstorms (frequency and the background environment in which they occur), their structure, propagation, and attendant severe weather for much of the West. However, understanding and monitoring thunderstorms in the western United States is becoming more important because of the significant increases in population and outdoor recreational activities.

Forecasters in the West will be more involved in monitoring and warning for thunderstorms as a result of the changes discussed above. We are conducting several thunderstorm studies using WSR-88D data from central Arizona. During the course of these studies, we have identified a number of limitations regarding the use of WSR-88D cell-based storm attributes and trends. This paper presents these initial findings.

2. Background

During studies conducted as part of the Southwest Area Monsoon Project (Howard et al. 1993; Maddox et al. 1997), thunderstorm cell-based attributes and life cycle trends were examined for storms occurring over central Arizona. The data used were from KIWA (the WSR-88D located at William’s Air Park near Phoenix, Arizona) during the summers of 1993–95. The examination of thunderstorm life cycles for more than 1100 separate storm cells showed that 1) storms in central Arizona typically exhibit short-lived single-cell characteristics, and 2) radar sampling and/or algorithm limitations can result in unreliable life cycle characteristics (i.e., radar-derived, cell-based attribute trends).

Complex terrain around the KIWA radar causes beam blockage at the lowest elevation scans (0.5° and 1.5°) for nearly 50 radials. Thus, the detection of storm reflectivity and velocity structures, even within close range of the radar, can be greatly restricted. The missing information can lead to unrealistic determinations of storm cell attributes. In addition to radar blockage and relatively short life cycles, there is a tendency for thunderstorm cells in central Arizona to develop in close proximity to one another. Close clustering of cells appears to hamper the ability of the cell identification algorithm to associate storm cells correctly from one radar volume scan to those present in the next volume scan. If this occurs, cell-based trends do not reflect the real situation. Because we noted unusual life cycle characteristics (i.e., radar-derived, cell-based attribute trends) that were different from what one might expect (i.e., decreasing radar tops with increased lightning frequency) for a large number of storms over central Arizona, we felt we needed to examine the measurement uncertainties introduced because of the character of the radar and scanning strategies used operationally by the NWS.

3. Methodology

A plot of 450 storm-top heights (defined as the maximum height of the 30-dBZ core; see NOAA 1991) determined by the KIWA radar for 20 August 1994 is shown in Fig. 2. The storm-top height measurements are aligned along the center of the beam axes for combined volume coverage pattern-21 (VCP-21) and VCP-11 elevation scans (Fig. 3, see reference). Although the WSR-88D algorithm determines the storm-top height computed to be along the beam centerline, the actual storm top may be higher. The uncertainty associated with storm-top height measurement is shown in Figs. 4a and 4b. For example, Fig. 4a shows that at 100-km range from the radar, the uncertainty between the radar-observed storm top and actual storm-top height can be greater than 3000 m for very tall storms (i.e., storms tops above 10 km).

The data presented in Fig. 4 are also applicable to other radar-based height measurements (e.g., height of maximum reflectivity, cloud top, storm base, etc.). It is important to note that when the 30-dBZ echo tops are located in regions of high gradients of uncertainty, small horizontal changes in the location of the cell can lead to large changes in the measured echo-top height. For example, a cell with 30-dBZ echo top of 10 km at 55-km range when scanned by a WSR-88D operating in VCP-21 (see Fig. 4a) will have a measured storm top of almost exactly 10 km. However, if the storm-top height remains constant and the cell’s range increases just enough so that the 9.9° elevation scan does not detect the 30-dBZ echo, the next radar volume scan would indicate that the 30-dBZ echo top has fallen to about 6 km. The trend of the echo-top height derived from the two consecutive volume scans would indicate a rapidly decreasing storm top; whereas, in this simple example the actual top of the 30-dBZ echo would not have changed. Operating the WSR-88D in the VCP-11 mode greatly diminishes the magnitude of the uncertainty inherent in the measurement of “heights.” It is important that the user of the operational radar system understand the magnitude of the inherent uncertainties in height measurements; however, the user also needs to understand that the horizontal movement of cells can produce large “trends” that may or may not be caused by actual storm evolution.

A simple computer program was used to simulate idealized thunderstorm life cycles and the radar depiction of the idealized storm as measured by the WSR-88D. We used this process to examine the evolution of the maximum height of the 30-dBZ core as detected by two of the operational WSR-88D scanning modes (VCP-11 and VCP-21). The program computes the differences of “radar-observed” heights versus actual heights of the idealized 30-dBZ core top. The program provides a simple way to assess the inherent uncertainties of the WSR-88D radar-observed 30-dBZ core height (remember that this attribute is defined as the “storm top”). To assess the WSR-88D’s limitations to detect storm top and associated trend, we varied, within our simple model, storm cell duration, storm cell distance from the radar, and storm cell maximum top height.

The storm-top height of the model cell is determined using the following assumptions: 1) the 30-dBZ core fills at least half the depth of the beam volume, 2) the storms are occurring in a standard atmosphere (i.e., index of refraction = 1.33), 3) storm-top height above ground level includes effects of earth curvature but ignores local terrain variability, 4) there are no local terrain features that block the lowest elevations, and 5) there are no data ambiguities due to “second trip echoes,” attenuation, or sidelobe contamination.

The idealized reflectivity structure we used depicts the life cycle of a “pulse”-type, single-cell thunderstorm (Doswell 1985). Pulse-type storms are common in moderately unstable, weakly sheared environments and have short-lived, strong updrafts. We focus primarily on this type of storm due to their frequency in central Arizona and because they often produce severe, short-lived downburst winds, an operational warning challenge. After the storm location and its movement are specified, the model computes the storm top that would be detected by the WSR-88D through the entire life cycle of the idealized cell. Figure 5 shows both the evolution of the reflectivity structure and the top of the 30-dBZ core for an idealized pulse-type storm. The “detected” storm-top heights are compared to the actual 30-dBZ core-top heights of the modeled cell.

4. Results

Using the model pulse storm reflectivity structure described in section 3, trend plots of the top of the 30-dBZ echo core were constructed for VCP-11 and VCP-21 scanning modes, for cells located at various distances, and for varying storm speeds toward the radar. The initial idealized storm is 8 km high at 7 min, reaches a maximum height of 14.3 km at 21 min, and then descends to 3.4 km at 42 min. The storm duration was defined to be 42 min (see Fig. 5). “Radar-determined” 30-dBZ echo top height trend plots were simulated for cells initially located at ranges of 25, 50, 100, 150, 200, and 250 km from the radar. The cells were defined to be stationary or to move toward the radar at 5 or 10 m s−1. Examples of individual trend plots are shown in Figs. 6–10. The uncertainties in simulated radar height measurements, essentially the difference between the idealized storm top and the radar-determined storm top, are referred to simply as “errors” for this paper. Since several large metropolitan areas are located 50–100 km from the nearest WSR-88D(s) (e.g., Phoenix, Albuquerque, Denver, Dallas, etc.), storm-top trends for cells initially located at 50- and 100-km range are discussed in the following paragraphs.

Figures 6a–c show trend plots for the idealized storm initially developing at 100-km range. With a stationary storm (Fig. 6a), radar-determined storm top is characterized by a relatively flat trend in storm-top height during the critical portion of the storm’s life cycle (i.e., from 7 to 21 min). Only during the initial and termination stages of the storm’s life cycle does the radar-determined trend agree well with the evolution of the model storm top. With a 5 m s−1 movement toward the radar (Fig. 6b), radar-determined trends indicate a descent in the storm top during the period the storm is actually growing to its maximum storm-top height. Figure 6c shows that this false trend is amplified when the model echo moves with increased velocity toward the radar. Similar trend plots for storm tops of cells initiating at 50-km range (Figs. 7a–c) produce similar life cycle errors. For a storm 50 km from the radar (Fig. 7a) the overall life cycle of the model stationary storm is captured reasonably well. When this cell moves at 5 m s−1 toward the radar, the trend indicates a decaying storm during the critical period from 15 to 25 min. A 10 m s−1 movement (Fig. 7c) toward the radar results in a trend showing rapid vertical growth but with substantial underestimation of the model’s real storm-top height. After 20 min, the radar detects the model cell’s evolution (decay) quite accurately. For similar model cell locations and movements, the WSR-88D VCP-11 scanning mode substantially improves the ability of the radar to detect the actual trends (e.g., Figs. 8a–c). If the modeled cell is located initially at 150-km range, both scanning modes (VCP-11 and -21) capture the evolution of the storm reasonably well but also substantially underestimate the storm-top height (see Figs. 9a and 9b).

Severe errors in measuring storm-top heights occur with storms very near the radar, where the maximum elevation angle of the 19.5° tilt severely restricts the monitoring capabilities of the operational radar. Forecasters are aware that storms whose tops grow into the “cone of silence” (defined as the region above the 19.5° tilt; see Fig. 3) cannot be reliably monitored; however, the impacts upon apparent storm trends may not be as obvious to the forecaster (especially periods of widespread severe storm activity). Figure 10a shows an example of detected storm-top height trend for the model storm moving into the cone of silence. The radar trends do not relate to the actual evolution of the model cell except at the beginning and at the end of its life cycle. Other situations that can lead to poor radar detection of the actual life cycle evolution of storm cells are storms moving into and through areas where the radar is blocked by man-made objects or by mountainous terrain, for very rapidly moving or evolving storm cells, and for cells at ranges greater than 150 km (e.g., cf. Fig. 10b and Fig. 9a). Many of these problem situations occur frequently in the western United States.

5. Conclusions

The simple model and approach used in this study provide useful tools for examining WSR-88D uncertainties detecting the 30-dBZ storm-top height and the inherent limitations of the radar-based depiction of thunderstorm evolution. The modeled “pulse” storm used illustrates the radar limitations well; however, the forecaster must deal with storms that are often behaving in much more complex ways. While the WSR-88D’s uncertainties are an inherent aspect of the sensing system, it is important that the user remain aware of the impacts of these uncertainties on radar-derived diagnostics. The uncertainties inherent in radar-derived trends for parameters such as storm top, height of maximum reflectivity, storm base, and cloud top can be substantial, as illustrated by the simple examples shown here.

From both an operational and a research perspective, these uncertainties must be considered when using WSR-88D observations to characterize thunderstorms and their trends. This is especially true in determining the interrelation between thunderstorms, their life cycle characteristics (i.e., rate of growth/decay), and storm-associated phenomena (lightning frequency, damaging winds, etc.) and severity. In some operational situations the trends of radar-derived storm features, such as storm top, can be reasonably accurate and of significant use to forecasters (Johnson et al. 1995); however, in other situations the trends can be incorrect and potentially confusing. Clearly, caution is required on the part of the user. We feel that a more realistic model should be developed for organized storm types and be used to evaluate the reliability of radar-derived trends for long-lived, fast-moving severe thunderstorms, that is, for storm types that frequently bring severe weather to the central and eastern United States.

Acknowledgments

We are grateful to Kurt Hondl and other group members of the NSSL Storm Research and Analysis Division for their valuable discussions and access to the cell attribute information. The comments of Don Burgess have been most useful to the authors. Portions of the research reported here have been supported by the Salt River Project and the Electric Power Research Institute under Contract W03978-02.

REFERENCES

  • Crum, T. D., and R. L. Alberty, 1993: The WSR-88D and the WSR-88D operational support facility. Bull. Amer. Meteor. Soc.,74, 1669–1687.

    • Crossref
    • Export Citation
  • Doswell, C. A., III, 1985: The operational meteorology of convective weather. Vol. II: Storm scale analysis. NOAA Tech. Memo. ERL ESG-15, Boulder, CO, 102–103.

  • Eilts, M. D., and Coauthors, 1996: Severe weather warning decision support system. Preprints, 18th Conf. on Severe Local Storms, San Francisco, CA, Amer. Meteor. Soc., 536–540.

  • Howard, K. W., C. L. Dempsey, and D. M. McCollum, 1993: SW Area Monsoon Project: Daily operations summary. NSSL, 435 pp. [Available from National Severe Storms Laboratory, Norman, OK 73069.].

  • Johnson, J. T., and Coauthors 1995: Operational testing of enhanced WSR-88D algorithms and display concepts in National Weather Service offices. Preprints, 27th Conf. on Radar Meteorology, Vail, CO, Amer. Meteor. Soc., 170–172.

  • Maddox, R. A., C. L. Dempsey, and K. W. Howard, 1997: Intense convective storms with little or no lightning over central Arizona—A case of inadvertent weather modification? J. Appl. Meteor.,36, 302–314.

    • Crossref
    • Export Citation
  • NOAA, 1991: Doppler radar meteorological observations, Part C, WSR-88D products and algorithms. Federal Meteorological Handbook, Office of the Federal Coordinator for Meteorological Services and Supporting Research, FCH-H11C-1991, Rockville, MD, 210 pp.

  • Witt, A., and J. T. Johnson, 1993: Operational testing of enhanced WSR-88D algorithms and display concepts in National Weather Service offices. Preprints, 26th Conf. on Radar Meteorology, Norman, OK, Amer. Meteor. Soc., 141–143.

Fig. 1.
Fig. 1.

Display from the WDSS system from the KIWA radar (see Eilts et al. 1996). The change with time of various parameters associated with the thunderstorm cell north of Globe, Arizona, is shown in the “trend” windows along the right side of the figure.

Citation: Weather and Forecasting 12, 1; 10.1175/1520-0434(1997)012<0166:UIWMAT>2.0.CO;2

Fig. 2.
Fig. 2.

A scatterplot of observed storm tops vs radar range for 450 identified storm cells from the KIWA, radar on 20 August 1994.

Citation: Weather and Forecasting 12, 1; 10.1175/1520-0434(1997)012<0166:UIWMAT>2.0.CO;2

Fig. 3.
Fig. 3.

(a) Radar beam geometry (range vs height) for VCP-11. Beamwidth is 0.95°, and there are 14 different elevation scans with a 5-min update rate, (b) same figure for VCP-21, which has only nine different elevation scans with a 6-min update rate (NOAA 1991).

Citation: Weather and Forecasting 12, 1; 10.1175/1520-0434(1997)012<0166:UIWMAT>2.0.CO;2

Fig. 4.
Fig. 4.

Uncertainty (m) between the WSR-88D determined height of a reflectivity target and the actual height for (a) VCP-21 and (b) VCP-11. The uncertainty of the height measurement is zero at the beam centerline, and the elevation angles of the various beam centerlines are indicated. Note that the height of features that do not reach the centerline of the 0.5° beam cannot be determined. The height of features located to the left of the centerline of the 19.5° beam cannot be determined.

Citation: Weather and Forecasting 12, 1; 10.1175/1520-0434(1997)012<0166:UIWMAT>2.0.CO;2

Fig. 5.
Fig. 5.

Vertical reflectivity structure during the life cycle of a pulse-type thunderstorm. Contours are 10-, 30-, and 50-dBZ reflectivity. The solid line denotes the evolution of the height of the top of the 30-dBZ echo core (i.e., WSR-88D storm top). Figure is after Doswell (1985).

Citation: Weather and Forecasting 12, 1; 10.1175/1520-0434(1997)012<0166:UIWMAT>2.0.CO;2

Fig. 6.
Fig. 6.

Plot of radar-observed storm top and the model storm top for VCP-21 for storm initiation at range 100 km (a) stationary, (b) movement toward the radar at 5 m s−1, and (c) movement toward the radar at 10 m s−1. Data points are at the end of each radar scan and are 6 min apart.

Citation: Weather and Forecasting 12, 1; 10.1175/1520-0434(1997)012<0166:UIWMAT>2.0.CO;2

Fig. 7.
Fig. 7.

Plot of radar-observed storm top and the model storm top for VCP-21 for storm initiation at range 50 km. Data points are at the end of each radar scan and are 5 min apart. Details are similar to Fig. 6.

Citation: Weather and Forecasting 12, 1; 10.1175/1520-0434(1997)012<0166:UIWMAT>2.0.CO;2

Fig. 8.
Fig. 8.

Plot of radar-observed storm top and the model storm top for VCP-11 for (a) stationary at range 50 km, (b) movement toward the radar at 5 m s−1 and at 50-km range, and (c) movement toward the radar at 5 m s−1 at range 100 km.

Citation: Weather and Forecasting 12, 1; 10.1175/1520-0434(1997)012<0166:UIWMAT>2.0.CO;2

Fig. 9.
Fig. 9.

Plot of the radar-observed storm top and the model storm top for (a) VCP-11 with movement toward the radar at 5 m s−1 and storm initiation at range 150 km, and (b) VCP-21 for the same storm.

Citation: Weather and Forecasting 12, 1; 10.1175/1520-0434(1997)012<0166:UIWMAT>2.0.CO;2

Fig. 10.
Fig. 10.

Plot of the radar-observed storm top and the model storm top for VCP-11 for (a) movement toward the radar at 5 m s−1 and storm initiation at range 25 km and (b) movement toward the radar at 5 m s−1 and storm initiation at range 200 km.

Citation: Weather and Forecasting 12, 1; 10.1175/1520-0434(1997)012<0166:UIWMAT>2.0.CO;2

Save