Date of Award

2015

Document Type

Thesis

Degree Name

Master of Science (MS)

Department

Atmospheric Science

Committee Chair

Lawrence D. Carey

Committee Member

Phillip M. Bitzer

Committee Member

John R. Mecikalski

Subject(s)

Radar meteorology, Rain and rainfall--Mathematical models, Radar meteorological stations

Abstract

Rainfall has spatial and temporal variability due to the microphysical properties of precipitation. Radar-rainfall algorithms are typically based on assumed raindrop and radar scattering models which include fitting a mathematical function such as a gamma model with three parameters (N0, mu, lamda) to the drop size distribution (DSD). Disdrometer observations are typically combined into a radar model to produce radar observables that are then used to derive empirical radar-rain algorithms. If radar and disdrometer-derived radar measurements are to be inter-compared, the appropriate spatial and temporal scales should be considered. For example, groundbased radar spatial resolutions are on the order of 300 meters while Global Precipitation Measurement Dual-Precipitation Radar has a footprint of 5 km which result in different temporal scales. Integration or sampling period is an important assumption in DSD measurement retrieval from disdrometer data. Prior studies have often assumed 1-minute disdrometer DSD samples based largely on trial-and-error empirical experience. No study to date has documented relative bias and standard error for radar-rain algorithms as a result of DSD integration period assumptions. Therefore, a series of experiments are conducted for various algorithms to evaluate and document bias and standard errors by assuming a truth integration period vs. model t integration period. Experimental results show attenuation estimated from specific differential phase at C-band has a -15% bias when 1 min truth integration is compared against 10 min integration modeled. Rain rate estimated from horizontal and differential reflectivity showed 8% bias when 10 min truth integration is compared against 1 min integration modeled. Uncertainty in integration period should be noted when validating satellite rainfall algorithms against ground-based disdrometer observations because of different implied spatial scales.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.