scispace - formally typeset
Search or ask a question
Author

Roderic L. Olsen

Bio: Roderic L. Olsen is an academic researcher. The author has an hindex of 1, co-authored 1 publications receiving 117 citations.

Papers
More filters
Journal ArticleDOI
TL;DR: In this article, the authors describe the modeling of the average raindrop size distribution measured at Ile-Ife, a tropical station in South Western Nigeria, using a method of moment regression, which allows a single lognormal function to be accurately fitted over the entire range of rain rates measured, a convenient result for rain attenuation and scattering applications.
Abstract: The paper describes the modeling of the average raindrop size distribution measured at Ile-Ife, a tropical station in South Western Nigeria. The new “method of moment regression” employed is a systematic one that has allowed a single lognormal function to be accurately fitted over the entire range of rain rates measured, a convenient result for rain attenuation and scattering applications. The Marshall and Palmer law, frequently used for such calculations, has been shown to be inadequate as a model for this tropical station, especially at the high rain rates important for communications systems design. This result suggests that the similar Laws and Parsons dropsize distribution currently adopted by the International Radio Consultative Committee for estimating rain attenuation from 1 to 1000 GHz may not be adequate over such a large frequency range in such tropical locations. Some comparisons are also made with the lognormal models and modeling approaches of other workers. The results confirm the utility of the rain rate parameterization used in the model for Ile-Ife, while giving further support to the accuracy of the lognormal distribution.

123 citations


Cited by
More filters
Journal ArticleDOI
TL;DR: In this article, a framework based on measured raindrop size distribution (DSD) data has been developed to assess uncertainties in DSD models employed in Ku-and Ka-band dual-wavelength radar retrievals.
Abstract: A framework based on measured raindrop size distribution (DSD) data has been developed to assess uncertainties in DSD models employed in Ku- and Ka-band dual-wavelength radar retrievals. In this study, the rain rates and attenuation coefficients from DSD parameters derived by dual-wavelength algorithms are compared with those directly obtained from measured DSD spectra. The impact of the DSD gamma parameterizations on rain estimation from the Global Precipitation Measurement mission (GPM) Dual-Frequency Precipitation Radar (DPR) is examined for the cases of a fixed shape factor μ as well as for a constrained μ—that is, a μ–Λ relation (a relationship between the shape parameter and slope parameter Λ of the gamma DSD)—by using 11 Particle Size and Velocity (Parsivel) disdrometer measurements with a total number of about 50 000 one-minute spectra that were collected during the Iowa Flood Studies (IFloodS) experiment. It is found that the DPR-like dual-wavelength techniques provide fairly accurate est...

76 citations

Journal ArticleDOI
TL;DR: In this article, a review of the deficiencies and their likely impact on system performance is presented, highlighting difficulties and encouraging realistic expectations and examining the question of whether costly, application-specific measurement programs might now be avoided by mining increasingly sophisticated national meteorological records.
Abstract: Due to pressures in the commercial sphere for increased information transfer rates or from the military to further miniaturize equipment to improve its portability, there has been a steady move toward higher carrier frequencies. For more than a decade now, increasing numbers of satellite communications systems have been in operation at millimeter wavelengths. Even where there is a clear view of the satellite from the ground station, a range of meteorological phenomena still combine to make propagation impairment a serious problem. Because of their complexity, difficulties in modeling them, and deficiencies in basic physical understanding, progress in engineering such systems has involved a large dose of empiricism. Ideally, this means a solid database of propagation measurements, made as nearly as possible in the intended geographical area of operations. Because of the time and cost of their accumulation, attempts have been made to use classification into climate zones on a worldwide basis to extend existing data to a wider range of situations. However, most of the measurements in the databases have been made in northern hemisphere temperate zones. Much evidence suggests that in tropical and equatorial zones, the factors that make important contributions to propagation impairment are different, putting a strong question mark over the use of existing design methods. This review looks at those deficiencies and their likely impact on system performance. At this time, rather than providing definitive answers - which must wait on further work - the best that can be done is to highlight difficulties and encourage realistic expectations. In a final section, the review examines the question of whether costly, application-specific measurement programs might now be avoided by mining increasingly sophisticated national meteorological records.

71 citations

Journal ArticleDOI
TL;DR: In this article, the results of the modeling of drop size distributions (DSD) observed during a 2-yr study in Barcelona are presented. But the results are limited to three standard distribution functions: exponential, gamma, and lognormal.
Abstract: This study shows the results of the modeling of drop size distributions (DSD) observed during a 2-yr study in Barcelona. Thirty-second individual sample collections of drop sizes and velocities were measured with an optical disdrometer and grouped into different classes according to their rain rate. Using the moments method, the entire experimental dataset was fitted to three standard distribution functions: exponential, gamma, and lognormal. Relationships were found between rain rate R and other moments of the DSD, such as optical attenuation Σ, liquid water content W, and reflectivity Z. Although gamma distribution generally reproduced experimental measurements more accurately, the Z(R) relationship, which is particularly relevant in radar meteorology, yielded the best results when calculated from fitted exponential distribution.

57 citations

Journal ArticleDOI
TL;DR: In this paper, the authors used the method of moments to estimate the parameters of lognormal DSD and optimized the modeled DSD parameters by examining the root mean square (RMS) error and the average probability ratio (APR) in estimation of the rain rate, rain attenuation, and radar reflectivity factor simultaneously.
Abstract: As communication services using higher frequencies are growing rapidly in the tropics, there is an increasing need for a finer model to predict the attenuation due to rain. The raindrop size distribution (DSD) is one of the major sources of error in any prediction model, mainly because of its variability in both space and time. The DSD parameters are computed from distrometer data that are classified into stratiform and convective types using S-band radar data. The method of moments is employed to estimate the parameters of lognormal DSD. The modeled DSD parameters are optimized by examining the root mean square (RMS) error and the average probability ratio (APR) in estimation of the rain rate, rain attenuation, and radar reflectivity factor simultaneously. The proposed model gives maximum (close to unity) APR and minimum RMS error when compared to any other set of DSD parameters.

53 citations

Journal ArticleDOI
TL;DR: In this article, the distribution of pluies in trois lieux of Nigeria (Calabar, Ile-Ife, and Zaria) has been investigated, i.e., bruine, averse, pluie etendue, orage.
Abstract: Les distributions des tailles des gouttes de pluie ont ete mesurees en trois lieux du Nigeria (Calabar, Ile-Ife et Zaria). Les donnees collectees pendant trois ans ont servi a obtenir des modeles de ces distributions pour divers types de pluie : bruine, averse, pluie etendue, orage. La distribution lognormale constitue un bon ajustement pour tous les types de pluie en des lieux tropicaux, sauf pour la bruine ou le modele exponentiel convient. On discute aussi les consequences de ces modeles pour des calculs de l’affaiblissement lineique du aux pluies tropicales.

42 citations