Search Results

You are looking at 1 - 4 of 4 items for :

  • Author or Editor: Peter S. Ray x
  • Journal of Atmospheric and Oceanic Technology x
  • Refine by Access: All Content x
Clear All Modify Search
Terence Given
and
Peter S. Ray

Abstract

The wind field resulting from a two-dimensional dual-Doppler synthesis algorithms is spectrally modified from the true wind field. The effects of spatial filtering on wind fields from the processes of interpolation, the averaging of pulses, and the effect of the finite radar pulse dimension were assessed. The effect resulting from the use of different interpolation techniques was also evaluated. Of those techniques tested, the best are the Cressman distance-weighted averaging and linear distance-weighted averaging, with the closest neighbor and uniform weighting having more undesirable characteristics.

The optimum influence radius is defined as the influence radius at which the ratio of the rms difference between the Fourier and least-squares responses (a measure of the aliasing) and the variance of the filtered wind field is minimized. This seeks to minimize the effect of energy aliased into scales other than the input wavelength. For the Cressman interpolation technique, the optimum influence radius is between 1.85 and 2.25 times the maximum data spacing. The range of acceptable influence radii includes consideration of the filtering by the radar of the data as it is collected, as well as the resolution of the final dataset. The optimum influence radius is dependent upon the largest data separation in the analysis domain. The absolute optimum influence radius is not significantly affected by inclusion of the radar-beam filtering effects.

Full access
Peter S. Ray
and
Mary Stephenson

Abstract

On 20 April 1984, the NOAA WP-3D aircraft, equipped with a Doppler radar in its tail, flew around a growing thunderstorm new Norman, Oklahoma. Doppler wind data was collected as the airplane flew six legs around the storm. During this time, the National Severe Storms laboratory (NSSL) dual-Doppler network collected data on the same storm. Different combinations of synthesis techniques were examined employing direct and pseudo-dual-Doppler observations from aircraft alone, and combinations of aircraft and ground-based Doppler radar. The effect of temporal resolution errors was assessed and related to uncertainties caused by geometric configuration. For this system, it was found that although the aircraft did provide useful data by extending the analysis to the region between the ground-based radars, the contribution was limited by the rapid evolution of the storm. Greater utility may generally be found for storms that evolve less rapidly.

Full access
Peter S. Ray
and
David P. Jorgensen

Abstract

Observations with airborne Doppler radar can expand the area of coverage and extend the time a moving weather system can remain under observation. Also, additional analysis methods are possible with the increase in independent estimates of the wind field that can be provided by an airborne sampling system. However, the advantages of airborne Doppler sensing are constrained by the geometry in which the data are collected, as well as errors introduced by uncertainties in the sampling platform location and orientation. Finally, a longer time required to sample a region than is typical for ground-based radar results in increased uncertainties due to the field's evolution and advection during the sampling interval. Uncertainties related to geometry are examined for flight patterns which are for aircraft alone and for those which also utilize data from one and two ground-based radars. These illustrate the distribution and relative magnitude of uncertainty expected for each type of flight pattern and data analysis method. Both the NOAA P−3, and the NCAR ELDORA scanning methodologies are examined.

To evaluate the different flight patterns, a relative quality index is used. It is defined as the reciprocal of the vertical velocity error variance integrated over the analysis domain. This normalized relative quality index is a mean value over the sampled volume. Flight patterns that utilize a single ground-based radar provide coverage over ∼ ten times the area in about one-half the time, and with relative quality about ten times better than that by aircraft alone.

Data collection, particularly aircraft data collection, often involves real-time decision making, and storms frequently are not in an ideal location relative to fixed ground-based radars. The best operational decisions require knowledge of eventual synthesis capabilities and the location of the volume to be interrogated relative to those facilities. These concepts are illustrated in a case example. Airborne Doppler and ground-based radar synthesis results are compared and discussed.

Full access
Karen L. Sangren
,
Peter S. Ray
, and
Gene B. Walker

Abstract

Doppler velocity spectra collected at vertical incidence contain information on vertical air motions and drop-size distributions with high spatial and temporal resolution. In the past, the computational interdependence between vertical air velocities and drop-size distributions has severely limited the accuracy with which they could be estimated. A dual-wavelength technique is applied in which vertical air motion is determined independently of the drop-size distribution. The Rogers reflectivity method and an extended version of the Hauser-Amayenc method are also applied. The latter technique fits Doppler spectra in a nonlinear least-squares sense using two exponential drop-size distribution models. Results of applying each method to Oklahoma squall line data are compared and the strengths and weaknesses of the three techniques are assessed. For the methods tested, there is a trade-off between potential accuracy and potential for successful application. For example, the dual wavelength method is theoretically quite accurate but is extremely sensitive to poor data quality.

Full access