Search Results

You are looking at 1 - 4 of 4 items for

  • Author or Editor: Albert J. Koscielny x
  • Refine by Access: All Content x
Clear All Modify Search
Albert J. Koscielny, R. J. Doviak, and R. Rabin

Abstract

Methods of statistical regression have been applied to single-radar radial velocity fields to map certain mesoscale (20–100 km) kinematic properties (e.g., divergence) of the convective boundary layer (CBL). Several methods, previously proposed, were found to produce estimates that were biased or whose variances were too large. When wind fields are linear on the meso- or larger scale, then single-Doppler velocity accuracies allow the estimation of horizontal divergence with an accuracy of about 4 × 10−5 s−1 and a resolution of ∼30 km, which may be sufficient to sense pre-thunderstorm convergence

A case study for 19 June 1980 suggests that single-Doppler weather radars of modest sensitivity can map the mesoscale divergence patterns within the cloud-free CBL. For this day, convergence zones to the northeast seem to precede cloud development by 1–2 h, and to the west precede thunderstorms by 3–4 h.

Full access
Albert J. Koscielny, Richard J. Doviak, and Dusan S. Zrnic

Abstract

Advances in clear air Doppler radar measurement have made practical the monitoring of radial velocities in the troposphere and lower stratosphere and even the vector wind, under some assumptions. Because the objective of wind profiling is to monitor winds representative of larger scale atmospheric motions, an assumption of a time-invariant spatially uniform wind field is commonly used. Then, the accuracy of the wind estimators depends on the error variance of the radial velocity, the departure from uniformity of the wind field and the measurement geometry.

We derive expressions for the variance and bias for some of these estimators when applied to a spatially linear wind field. The techniques we consider are three fixed beams, azimuthal scanning (VAD) and elevation scanning (VED). In addition, we examine a method based on the integration of the continuity equation to estimate the areal-averaged wind. This technique sometimes leads to better estimates than do direct methods.

Full access
Thomas R. Karl, Albert J. Koscielny, and Henry F. Diaz

Abstract

Principal component (PC) analysis performed on irregularly spaced data can produce distorted loading patterns. We provide an example to demonstrate some distorted patterns which can result from the direct application of PC analysis (or eigenvector analysis, factor analysis, or asymptotic singular decomposition) on irregularly spaced data. The PCs overestimate loadings in areas of dense data. The problem can be avoided by interpolating the irregularly spaced data to a grid which closely approximates equal-area.

Full access
Roger F. Reinking, Roger Caiazza, Robert A. Kropfli, Brad W. Orr, Brooks E. Martner, Thomas A. Niziol, Gregory P. Byrd, Richard S. Penc, Robert J. Zamora, Jack B. Snider, Robert J. Ballentine, Alfred J. Stamm, Christopher D. Bedford, Paul Joe, and Albert J. Koscielny

Snowstorms generated over the Great Lakes bring localized heavy precipitation, blizzard conditions, and whiteouts to downwind shores. Hazardous freezing rain often affects the same region in winter. Conventional observations and numerical models generally are resolved too coarsely to allow detection or accurate prediction of these mesoscale severe weather phenomena. The Lake Ontario Winter Storms (LOWS) project was conducted to demonstrate and evaluate the potential for real-time mesoscale monitoring and location-specific prediction of lake-effect storms and freezing rain, using the newest of available technologies. LOWS employed an array of specialized atmospheric remote sensors (a dual-polarization short wavelength radar, microwave radiometer, radio acoustic sounding system, and three wind profilers) with supporting observing systems and mesoscale numerical models. An overview of LOWS and its initial accomplishments is presented.

Full access