• Burgess, D. W., , and Crum T. D. , 2009: Observed failure modes of the WSR-88D velocity dealiasing algorithm during severe weather outbreaks. Preprints, 34rd Conf. on Radar Meteorology, Williamsburg, VA, Amer. Meteor. Soc., P5.16. [Available online at https://ams.confex.com/ams/pdfpapers/156056.pdf.]

  • Eilts, M. D., , and Smith S. D. , 1990: Efficient dealiasing of Doppler velocities using local environment constraints. J. Atmos. Oceanic Technol., 7, 118128.

    • Search Google Scholar
    • Export Citation
  • Gong, J., , Wang L. , , and Xu Q. , 2003: A three-step dealiasing method for Doppler velocity data quality control. J. Atmos. Oceanic Technol., 20, 17381748.

    • Search Google Scholar
    • Export Citation
  • Jing, Z., , and Wiener G. , 1993: Two-dimensional dealiasing of Doppler velocities. J. Atmos. Oceanic Technol., 10, 798808.

  • Liu, S., , Xu Q. , , and Zhang P. , 2005: Quality control of Doppler velocities contaminated by migrating birds. Part II: Bayes identification and probability tests. J. Atmos. Oceanic Technol., 22, 11141121.

    • Search Google Scholar
    • Export Citation
  • Liu, S., and Coauthors, 2009: WSR-88D radar data processing at NCEP. Preprints, 34th Conf. on Radar Meteorology, Williamsburg, VA, Amer. Meteor. Soc., 14.2. [Available online at https://ams.confex.com/ams/pdfpapers/156011.pdf.]

  • Tabary, P., , Scialom G. , , and Germann U. , 2001: Real-time retrieval of the wind from aliased velocities measured by Doppler radars. J. Atmos. Oceanic Technol., 18, 875882.

    • Search Google Scholar
    • Export Citation
  • Witt, A., , Brown R. A. , , and Jing Z. , 2009: Performance of a new velocity dealiasing algorithm for the WSR-88D. Preprints, 34rd Conf. on Radar Meteorology, Williamsburg, VA, Amer. Meteor. Soc., P4.8. [Available online at https://ams.confex.com/ams/34Radar/techprogram/paper_155951.htm.]

  • Xu, Q., , and Nai K. , 2013: A two-step variational method for analyzing severely aliased radar velocity observations with small Nyquist velocities. Quart. J. Roy. Meteor. Soc., in press.

    • Search Google Scholar
    • Export Citation
  • Xu, Q., , Nai K. , , Wei L. , , Zhang P. , , Zhao Q. , , and Harasti P. R. , 2009a: A real-time radar wind data quality control and analysis system for nowcast application. Extended abstracts, Int. Symp. on Nowcasting and Very Short Range Forecasting (WSN09), Whistler, British Columbia, Canada, WMO, 3.5. [Available online at http://www.nowcasting2009.ca/images/stories/Presentations/WSN09_Presentations.htm.]

  • Xu, Q., , Nai K. , , Wei L. , , and Zhao Q. , 2009b: An unconventional approach for assimilating aliased radar radial velocities. Tellus, 61A, 621630.

    • Search Google Scholar
    • Export Citation
  • Xu, Q., , Nai K. , , Zhang P. , , Liu S. , , and Parrish D. , 2009c: A new dealiasing method for Doppler velocity data quality control. Preprints, 34th Conf. on Radar Meteorology, Williamsburg, VA, Amer. Meteor. Soc., P9.6. [Available online at https://ams.confex.com/ams/pdfpapers/155947.pdf.]

  • Xu, Q., , Nai K. , , and Wei L. , 2010: Fitting VAD wind to aliased radial-velocity observations – A minimization problem with multiple minima. Quart. J. Roy. Meteor. Soc., 136, 451461.

    • Search Google Scholar
    • Export Citation
  • Xu, Q., , Nai K. , , Wei L. , , Zhang P. , , Liu S. , , and Parrish D. , 2011: A VAD-based dealiasing method for radar velocity data quality control. J. Atmos. Oceanic Technol., 28, 5062.

    • Search Google Scholar
    • Export Citation
  • Zhang, P., , Liu S. , , and Xu Q. , 2005: Quality control of Doppler velocities contaminated by migrating birds. Part I: Feature extraction and quality control parameters. J. Atmos. Oceanic Technol., 22, 11051113.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Flowcharts for the first two steps of the improved reference check: (a) the modified AR-VAD analysis performed in the first step to produce a first-guess radial-velocity field, denoted by , on each qualified circle from the highest to the second lowest tilt. (b) The refined AR-Var analysis performed in the second step to produce a reference radial-velocity field on each reference circle from the highest to the lowest tilt. The semiqualified circle searched and selected from each vertical layer (within zn ± ∆z/2 with ∆z = 25 m) on each tilt in the first step must satisfy the two data-coverage conditions described in section 3b of XN12. In the first step, a semiqualified circle is upgraded to a qualified circle only if has the required accuracy [(11) of XN12] and thus is accepted as a first guess. In the second step, a qualified (or semiqualified) circle is upgraded to a reference circle only if the refined AR-Var analysis can produce a radial-velocity field, denoted by , with an improved accuracy over the first guess (or proxy first guess) [see the last paragraph of section 2c and second paragraph of section 3b of XN12] and thus is accepted as a reference radial-velocity field, denoted by .

  • View in gallery

    (a) Raw level-II Doppler radial-velocity image at θ = 1.5° from KTLX radar at 0436:37 UTC 27 Jan 2009. (b) Reference radial-velocity field at θ = 1.5° produced by the refined AR-Var analysis in the first main step of the improved method. (c) Dealiased radial-velocity image at θ = 1.5° produced by the improved reference check in the first main step. (d) Dealiased radial-velocity image at θ = 1.5° produced by the final (second) main step of the improved method. Note that the reference radial-velocity field is a continuous field interpolated between reference circles, so it has no data hole.

  • View in gallery

    Zonal and meridional VAD wind components, denoted by (u, υ), plotted as functions of height (z) above the radar site. The long dark (solid and dotted) curves plot the VAD wind (u, υ) produced by applying the classic VAD analysis to the dealiased radial-velocity data (produced by the improved method). The short gray (solid and dashed) curves plot the VAD wind (u, υ) produced by applying the AR-VAD analysis directly to the aliased raw radial-velocity data for the case in Fig. 2.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 11 11 5
PDF Downloads 17 17 15

An Adaptive Dealiasing Method Based on Variational Analysis for Radar Radial Velocities Scanned with Small Nyquist Velocities

View More View Less
  • 1 NOAA/National Severe Storms Laboratory, Norman, Oklahoma
  • | 2 Cooperative Institute for Mesoscale Meteorological Studies, University of Oklahoma, Norman, Oklahoma
© Get Permissions
Full access

Abstract

Previous velocity–azimuth display (VAD)-based methods of dealiasing folded radial velocities have relied heavily on the VAD uniform-wind assumption and, thus, can fail when the uniform-wind assumption becomes poor around azimuthal circles in a vertical layer and the Nyquist velocity is small (≤12 m s−1). By using the two-step, alias-robust variational (AR-Var) analysis in place of the alias-robust VAD (AR-VAD) analysis for the reference check, the previous AR-VAD-based dealiasing method is improved to an AR-Var-based dealiasing method adaptively for radar radial velocities scanned with small Nyquist velocities. The method has been tested with severely aliased velocity data scanned by the Oklahoma KTLX radar. The robustness and satisfactory performance of the AR-Var-based dealiasing are exemplified by the results obtained for a severe winter ice storm scanned with the Nyquist velocity reduced to 11.51 m s−1.

Correspondence author address: Dr. Qin Xu, National Severe Storms Laboratory, 120 David L. Boren Blvd., Norman, OK 73072-7326. E-mail: Qin.Xu@noaa.gov

Abstract

Previous velocity–azimuth display (VAD)-based methods of dealiasing folded radial velocities have relied heavily on the VAD uniform-wind assumption and, thus, can fail when the uniform-wind assumption becomes poor around azimuthal circles in a vertical layer and the Nyquist velocity is small (≤12 m s−1). By using the two-step, alias-robust variational (AR-Var) analysis in place of the alias-robust VAD (AR-VAD) analysis for the reference check, the previous AR-VAD-based dealiasing method is improved to an AR-Var-based dealiasing method adaptively for radar radial velocities scanned with small Nyquist velocities. The method has been tested with severely aliased velocity data scanned by the Oklahoma KTLX radar. The robustness and satisfactory performance of the AR-Var-based dealiasing are exemplified by the results obtained for a severe winter ice storm scanned with the Nyquist velocity reduced to 11.51 m s−1.

Correspondence author address: Dr. Qin Xu, National Severe Storms Laboratory, 120 David L. Boren Blvd., Norman, OK 73072-7326. E-mail: Qin.Xu@noaa.gov

1. Introduction

In the dealiasing method of Xu et al. (2011, hereafter X11), the alias-robust velocity–azimuth display (AR-VAD) analysis (Xu et al. 2010, hereafter X10) was used to replace the modified VAD analysis (Tabary et al. 2001; Gong et al. 2003) for the reference check to produce critical alias-free data, called seed data, in the first main step. This AR-VAD-based dealiasing method has been tested with real-time observations from the KTLX radar and other operational Weather Surveillance Radar-1988 Doppler (WSR-88D) radars under various weather conditions. From these tests, the method is found to be capable of correcting alias error, without false dealiasing in most cases, but it occasionally fails to correct or flag severely aliased radial velocities around strongly sheared inversion layers (that are not perfectly flat in the horizontal) in severe winter ice storms scanned by the operational WSR-88D radars using volume coverage pattern 31 (VCP31) with the Nyquist velocity, denoted by υN, reduced below 12 m s−1. In the latter case, the estimated VAD wind often cannot pass the stringent threshold for vertical gaps, so the vertical profile of the VAD wind is severely limited below the inversion layer.

Since the AR-VAD analysis is inherently limited by the VAD uniform-wind assumption, its resultant reference radial-velocity field is purely sinusoidal and not flexible enough to represent the true field with the required accuracy. An alias error from a bad seed datum can propagate to a large area through the continuity check in the subsequent step. To ensure that no bad seed datum is produced by the reference check, the reference radial velocity must be within ±(2υN + ɛoN) ≈ ±(2 – q)υN of the true radial velocity at every datum point, where N is the threshold value for the reference check (with q = ¼ used in X11, but relaxed to q = 0.8 to enhance the seed data coverage in this paper) and |ɛo| ≪ υN is assumed for the random observation error ɛo. This required accuracy is hard to achieve at every datum point by the AR-VAD analysis when υN < 12 m s−1. The reduced υN in VCP31 has also been a difficult challenge for the operationally used dealiasing techniques (Eilts and Smith 1990; Jing and Wiener 1993; Burgess and Crum 2009; Witt et al. 2009).

The alias-robust variational (AR-Var) analysis (Xu et al. 2009b, hereafter X09b) is not limited by the VAD uniform-wind assumption, but it needs a first-guess radial-velocity field with a required accuracy that is tied up with υN (although not as stringent as that aforementioned for the reference field). Such a first guess was obtained from a numerical weather prediction (NWP) model for the example in section 3.2 of X09b, but an NWP model does not always have the required accuracy especially when υN < 12 m s−1. To overcome this limitation, Xu and Nai (2012, hereafter XN12) developed a two-step AR-Var analysis by relaxing the VAD uniform-wind assumption and modifying the AR-VAD analysis to produce a first guess with improved flexibility and accuracy in the first step, so the AR-Var analysis can be refined and performed in the second step without needing a first guess from a NWP model. This two-step AR-Var analysis can replace the AR-VAD analysis to improve the reference check and, therefore, yields an improved AR-Var-based dealiasing method for VCP31. Other than the improved reference check, the AR-Var-based dealiasing performs the same preprocessing step and the same continuity check in the second main step as in X11. The improved reference check is described in the next section. An example is presented in section 3 to show the performance of the method. Conclusions follow in section 4.

2. Improved reference check

The improved reference check performs three steps: 1) the modified AR-VAD analysis (i.e., the first step of the two-step AR-Var analysis in section 2b of XN12) to produce a first-guess radial-velocity field, denoted by , on each qualified circle from the highest to the second lowest tilt; 2) the refined AR-Var analysis (i.e., the second step of the two-step AR-Var analysis in section 2c of XN12) to produce a reference radial-velocity field, denoted by , on each reference circle from the highest to the lowest tilt; and 3) the reference check to produce seed data in two substeps. The term “circle” refers to a range ring of constant radial distance, while the term “tilt” refers to an elevation angle or a sweep of radar scan at that elevation angle. The algorithm flowcharts for steps 1–2 are plotted in Fig. 1. The two substeps in step 3 are described below:

  1. Linearly interpolating from the reference circles to the remaining circles along each radar beam to obtain the reference radial-velocity field.
  2. Perform the reference check as described in section 2b of X11 except that (i) is now produced by the refined AR-Var analysis (step 2) instead of the AR-VAD analysis, (ii) the coverage is no longer limited by the cutoff radial range (r = 30 km for θ ≈ 0.5° or r = 80 km for θ ≥ 1°), and (iii) the threshold value [(4)-(5) of X11] is relaxed from υN/4 to 0.8υN (so must be within ±1.2υN of the true υr as explained in the introduction). In addition, the accepted datum must not only pass the above threshold, but also have the same sign as . This sign consistency condition can avoid occasional false dealiasing caused by the relaxed threshold (0.8υN) when the raw radial velocities become extremely noisy around zero over a relatively broad area.
    Fig. 1.
    Fig. 1.

    Flowcharts for the first two steps of the improved reference check: (a) the modified AR-VAD analysis performed in the first step to produce a first-guess radial-velocity field, denoted by , on each qualified circle from the highest to the second lowest tilt. (b) The refined AR-Var analysis performed in the second step to produce a reference radial-velocity field on each reference circle from the highest to the lowest tilt. The semiqualified circle searched and selected from each vertical layer (within zn ± ∆z/2 with ∆z = 25 m) on each tilt in the first step must satisfy the two data-coverage conditions described in section 3b of XN12. In the first step, a semiqualified circle is upgraded to a qualified circle only if has the required accuracy [(11) of XN12] and thus is accepted as a first guess. In the second step, a qualified (or semiqualified) circle is upgraded to a reference circle only if the refined AR-Var analysis can produce a radial-velocity field, denoted by , with an improved accuracy over the first guess (or proxy first guess) [see the last paragraph of section 2c and second paragraph of section 3b of XN12] and thus is accepted as a reference radial-velocity field, denoted by .

    Citation: Journal of Atmospheric and Oceanic Technology 29, 12; 10.1175/JTECH-D-12-00145.1

As the improved reference check can produce seed data to cover nearly the entire radial range (with no cutoff), the block-to-point continuity check (section 2c of X11) can be performed reliably in the second main step as long as it does not go beyond the radial range covered by the reference check. When υN < 15 m s−1, the block-to-point continuity check can become occasionally unreliable going into the far range where no seed datum is produced by the reference check. This presents a difficult challenge for the block-to-point continuity check or any types of continuity checks including those operationally used and currently developed at the NWS Radar Operations Center (Burgess and Crum 2009; Witt et al. 2009; Xu et al. 2009c). This difficulty is avoided as the improved reference check can cover almost the entire radial range (with no cutoff) in the first main step.

3. Performance of the method exemplified by an ice storm case

Figure 2a shows the raw radial velocities scanned at 1.5° tilt with VCP31 from the KTLX radar for the Oklahoma winter ice storm at 0436:37 UTC 27 January 2009. The raw radial velocities were severely aliased in most areas outside the 25-km radial range and the aliased velocities were even folded twice in areas to the southwest and northeast of the radar outside the 100-km radial range. Figure 2b shows the reference radial-velocity field produced by the AR-Var-based dealiasing for the case in Fig. 2a. This reference field covers the radial range up to r = 205 km (or a vertical range up to z = 7.9 km on 1.5° tilt), and it shows that the wind field is dominated by a strong northeasterly flow below the vertical level of z ≈ 0.8 km (or r ≈ 26 km on 1.5° tilt) and then switches sharply to a strong southwesterly flow above z ≈ 0.8 km. This sharp reverse in wind direction is a real feature and is well captured by the two-step AR-Var analysis but not retrieved by the AR-VAD analysis. The failure of the AR-VAD analysis is caused by the reduced υN (=11.51 m s−1) and the increased discrepancy between the VAD-analyzed uniform wind and the true wind especially around the vertical shear layer that is not perfectly flat.

Fig. 2.
Fig. 2.

(a) Raw level-II Doppler radial-velocity image at θ = 1.5° from KTLX radar at 0436:37 UTC 27 Jan 2009. (b) Reference radial-velocity field at θ = 1.5° produced by the refined AR-Var analysis in the first main step of the improved method. (c) Dealiased radial-velocity image at θ = 1.5° produced by the improved reference check in the first main step. (d) Dealiased radial-velocity image at θ = 1.5° produced by the final (second) main step of the improved method. Note that the reference radial-velocity field is a continuous field interpolated between reference circles, so it has no data hole.

Citation: Journal of Atmospheric and Oceanic Technology 29, 12; 10.1175/JTECH-D-12-00145.1

To show the problem, the classic VAD analysis is applied to the dealiased velocity data (produced by the method in this paper) and the resultant VAD wind components, denoted by (u, υ), are plotted in Fig. 3 for comparison with those produced by applying the AR-VAD analysis directly to the raw velocity data. As shown, the AR-VAD analysis starts to produce suspicious VAD winds when z increases to 0.6 km and beyond, and then fails completely when z increases beyond 1 km. This presents a serious limitation for the use of the AR-VAD analysis, and this limitation has been seen similarly in other cases of winter ice storms scanned by using VCP31 with reduced υN. It is thus necessary to use the two-step AR-Var analysis in place of the AR-VAD analysis for the reference check to overcome the above limitation and solve the problem caused by the reduced υN.

Fig. 3.
Fig. 3.

Zonal and meridional VAD wind components, denoted by (u, υ), plotted as functions of height (z) above the radar site. The long dark (solid and dotted) curves plot the VAD wind (u, υ) produced by applying the classic VAD analysis to the dealiased radial-velocity data (produced by the improved method). The short gray (solid and dashed) curves plot the VAD wind (u, υ) produced by applying the AR-VAD analysis directly to the aliased raw radial-velocity data for the case in Fig. 2.

Citation: Journal of Atmospheric and Oceanic Technology 29, 12; 10.1175/JTECH-D-12-00145.1

The reference radial-velocity field in Fig. 2b also shows some rapid wind direction shifts (with height) at least in the following three local areas: (i) a relative large far-range local area to the southeast of the radar site within the azimuthal range of 130° < φ < 170° and the radial range of 170 km < r < 185 km, (ii) a middle-range local area to the northwest of the radar site within the azimuthal range of 300° < φ < 350° and the radial range of 70 km < r < 80 km, and (iii) a small near-range local area to the northwest of the radar site within the azimuthal range of 320° < φ < 290° and the radial range of 40 km < r < 45 km. Note that all these areas largely overlap the data-void areas as shown in Fig. 2a, so the rapid wind direction shifts in these areas are mainly spurious features caused by the irregular variations of data void sectors (segments on each reference circle) between neighboring reference circles (range gate to range gate) in these local areas. Since these spurious features are in data-void areas, they do not harm the reference check. The reference check is thus still ensured to be free of false dealiasing even around data void areas, as exemplified in Fig. 2c.

There is a total of 244 763 raw velocity data points on the 1.5° tilt in Fig. 2a, and 230 246 raw data points (94.1% of the total) are within the radial range (5 km ≤ r ≤ 205 km) covered by the reference radial-velocity field in Fig. 2b. Among the covered 230 246 raw data, 219 825 data points (95.5% of the covered) are dealiased or deflagged by the reference check with no false dealiasing or false deflagging as shown in Fig. 2c. The remaining 10 421 raw velocity data are flagged (blacked) in Fig. 2c (cf. the raw data image in Fig. 2a), and they are mainly in the gray areas of near-zero reference radial velocity as shown in Fig. 2b. Among the 10 421 flagged data, 4241 flagged data are further deflagged by the block-to-point continuity check in the second main step (section 2c of X11), so there are 224 039 dealiased or deflagged data (97.3% of all data covered) with no false dealiasing or false deflagging as shown in Fig. 2d.

4. Conclusions

In this paper, the VAD-based dealiasing method (X11) is improved to an AR-Var-based adaptive dealiasing method to overcome the difficulties caused by the reduced υN (<12 m s−1) in VCP31. The method has been tested with severely aliased velocity data collected from the KTLX radar using VCP31 and the Federal Aviation Administration’s (FAA’s) TOKC radar using scan Mode80 (with υN < 15 m s−1), and the results are summarized in Table 1. This method is compared with the previous VAD-based methods in Table 2. The real-time radar wind analysis system (Xu et al. 2009a) has been used to monitor the performance of the method by displaying the dealiased radial velocities and their produced wind field (including vertical profile of horizontal wind as shown in Fig. 3) in comparison with observations from other sources [e.g., the National Oceanic and Atmospheric Administration (NOAA) wind profiler in Purcell, Oklahoma]. The AR-Var-based dealiasing also outperforms the AR-VAD-based dealiasing for other VCPs (with υN > 15 m s−1) that did not fail the latter method, but it costs triple the computation time and thus is used only for VCP31 and Mode80.

Table 1.

Summary of results for the AR-Var-based dealiasing tested with severely aliased velocity data collected from the KTLX (WSR-88D radar) using VCP31 (with υN < 12 m s−1) and from the FAA’s TOKC (TDWR radar near the Oklahoma City Airport) using scan Mode80 (with υN < 15 m s−1). Columns 6 and 7 list the ranges of percentages of correctly dealiased or deflagged data points and remaining flagged data points with respect to the total number of raw data points in each volume, respectively, for all the volumes listed in column 4. The last column indicates that there is no falsely dealiased or deflagged datum (at any of the 105–106 data points) in each volume. As shown in column 6, the dealiased data coverage can be as low as 50% for a rainstorm. Note that the raw radial-velocity data collected from a rainstorm are often scattered and cover less areas than that covered by the data collected from an ice storm. The number of semiqualified circles (selected under the two data-coverage conditions described in section 3b of XN12) is thus often reduced and so is the dealiased data coverage for a rainstorm.

Table 1.
Table 2.

Comparisons of the AR-Var-based method developed in this paper with the two previous dealiasing methods: the AR-VAD-based method of X11 and the three-step method of Gong et al. (2003). The test scores listed in the last four rows are obtained for each volume (below z = 10 km) that contains the tilt selected in each case (see the referred figure). All three methods were originally developed for operational applications to radar data assimilation at NCEP, so they are completely free of human adjustment during the dealiasing procedure. As explained in X11, the block-to-point continuity check procedure was developed, in place of the original continuity check in the three-step method, to enhance the use of available seed data in a properly enlarged block area around each flagged data point that is being checked with multiple threshold conditions to avoid false dealiasing, so it can go through (or around) data holes that the original continuity check fails to pass or occasionally produces false dealiasing. When the AR-Var-based method is applied to scan modes (with υN > 18 m s−1) other than VCP31 (for the three cases listed in the last three rows), the first-guess radial-velocity field is simply produced by the AR-VAD analysis (instead of the modified AR-VAD analysis in section 2b of XN12) in the first step of the AR-Var analysis.

Table 2.

The AR-Var-based dealiasing has been incorporated into the radar data quality control package (Zhang et al. 2005; Liu et al. 2005) delivered to the National Centers for Environmental Prediction (NCEP) for operational tests (Liu et al. 2009; Xu et al. 2009c). For radar data assimilation applications at NCEP, the method is required to be absolutely free of false dealiasing. This has inevitably sacrificed the data coverage to a certain degree. The sacrificed (flagged) data coverage is often minor and marginal for the improved reference check that can cover nearly the entire radial range. However, for the VAD-based dealiasing method currently used for other VCPs, the reference check in the first main step is limited within the cutoff radial range to avoid false dealiasing, so isolated data areas away (by 40 gates and 5 beams or more) from the seed data areas produced by the reference check will remain flagged as they cannot be reached by the block-to-point continuity check in the second main step, as explained in X11. By using wind information from other sources, it is possible to find a new way to produce acceptable first guesses for the AR-Var analysis applied adaptively to isolated data areas to improve the data coverage. Continued effort is being undertaken in this direction.

Acknowledgments

The authors are thankful to the anonymous reviewers for their suggestions that improved the presentation of the paper, and to Dr. Shun Liu for testing the method at NCEP. The research was supported by NCEP-NSSL radar data quality control project and ONR Grant-N000141010778 to University of Oklahoma. Funding was also provided by NOAA/OAR under NOAA–University of Oklahoma Cooperative Agreement NA17RJ1227.

REFERENCES

  • Burgess, D. W., , and Crum T. D. , 2009: Observed failure modes of the WSR-88D velocity dealiasing algorithm during severe weather outbreaks. Preprints, 34rd Conf. on Radar Meteorology, Williamsburg, VA, Amer. Meteor. Soc., P5.16. [Available online at https://ams.confex.com/ams/pdfpapers/156056.pdf.]

  • Eilts, M. D., , and Smith S. D. , 1990: Efficient dealiasing of Doppler velocities using local environment constraints. J. Atmos. Oceanic Technol., 7, 118128.

    • Search Google Scholar
    • Export Citation
  • Gong, J., , Wang L. , , and Xu Q. , 2003: A three-step dealiasing method for Doppler velocity data quality control. J. Atmos. Oceanic Technol., 20, 17381748.

    • Search Google Scholar
    • Export Citation
  • Jing, Z., , and Wiener G. , 1993: Two-dimensional dealiasing of Doppler velocities. J. Atmos. Oceanic Technol., 10, 798808.

  • Liu, S., , Xu Q. , , and Zhang P. , 2005: Quality control of Doppler velocities contaminated by migrating birds. Part II: Bayes identification and probability tests. J. Atmos. Oceanic Technol., 22, 11141121.

    • Search Google Scholar
    • Export Citation
  • Liu, S., and Coauthors, 2009: WSR-88D radar data processing at NCEP. Preprints, 34th Conf. on Radar Meteorology, Williamsburg, VA, Amer. Meteor. Soc., 14.2. [Available online at https://ams.confex.com/ams/pdfpapers/156011.pdf.]

  • Tabary, P., , Scialom G. , , and Germann U. , 2001: Real-time retrieval of the wind from aliased velocities measured by Doppler radars. J. Atmos. Oceanic Technol., 18, 875882.

    • Search Google Scholar
    • Export Citation
  • Witt, A., , Brown R. A. , , and Jing Z. , 2009: Performance of a new velocity dealiasing algorithm for the WSR-88D. Preprints, 34rd Conf. on Radar Meteorology, Williamsburg, VA, Amer. Meteor. Soc., P4.8. [Available online at https://ams.confex.com/ams/34Radar/techprogram/paper_155951.htm.]

  • Xu, Q., , and Nai K. , 2013: A two-step variational method for analyzing severely aliased radar velocity observations with small Nyquist velocities. Quart. J. Roy. Meteor. Soc., in press.

    • Search Google Scholar
    • Export Citation
  • Xu, Q., , Nai K. , , Wei L. , , Zhang P. , , Zhao Q. , , and Harasti P. R. , 2009a: A real-time radar wind data quality control and analysis system for nowcast application. Extended abstracts, Int. Symp. on Nowcasting and Very Short Range Forecasting (WSN09), Whistler, British Columbia, Canada, WMO, 3.5. [Available online at http://www.nowcasting2009.ca/images/stories/Presentations/WSN09_Presentations.htm.]

  • Xu, Q., , Nai K. , , Wei L. , , and Zhao Q. , 2009b: An unconventional approach for assimilating aliased radar radial velocities. Tellus, 61A, 621630.

    • Search Google Scholar
    • Export Citation
  • Xu, Q., , Nai K. , , Zhang P. , , Liu S. , , and Parrish D. , 2009c: A new dealiasing method for Doppler velocity data quality control. Preprints, 34th Conf. on Radar Meteorology, Williamsburg, VA, Amer. Meteor. Soc., P9.6. [Available online at https://ams.confex.com/ams/pdfpapers/155947.pdf.]

  • Xu, Q., , Nai K. , , and Wei L. , 2010: Fitting VAD wind to aliased radial-velocity observations – A minimization problem with multiple minima. Quart. J. Roy. Meteor. Soc., 136, 451461.

    • Search Google Scholar
    • Export Citation
  • Xu, Q., , Nai K. , , Wei L. , , Zhang P. , , Liu S. , , and Parrish D. , 2011: A VAD-based dealiasing method for radar velocity data quality control. J. Atmos. Oceanic Technol., 28, 5062.

    • Search Google Scholar
    • Export Citation
  • Zhang, P., , Liu S. , , and Xu Q. , 2005: Quality control of Doppler velocities contaminated by migrating birds. Part I: Feature extraction and quality control parameters. J. Atmos. Oceanic Technol., 22, 11051113.

    • Search Google Scholar
    • Export Citation
Save