Maximum-Likelihood Retrieval of Modeled Convective Rainfall Patterns from Midlatitude C-Band Weather Radar Data

Abstract

A spatial characterization of the midlatitude mesoscale rain fields from C-band radar measurements is performed by means of a systematic analysis and modeling of convective rain-cell bidimensional shapes and spatial correlation. A large rainfall dataset that is derived from an operational C-band dual-polarized radar, which is placed in S. Pietro Capofiume near Bologna (Italy), has been collected and analyzed for this purpose. Different models of convective rainy horizontal structures are described and compared. Special attention is devoted to the consolidated unimodal models (or unimodal patterns) like GAUCELL with a Gaussian rain-rate profile and EXCELL with an exponential rain-rate profile, and the hybrid models like HYCELL and DEXCELL based on a proper combination of the previous unimodal models. The new hybrid model DEXELL, which is introduced here, is an extension of the HYCELL model, which is previously proposed in literature. A pixel-by-pixel model numerical integration is carried out in order to perform a homogeneous comparison between the rain-cell model and the measured features such as peak, average, root mean square, gradient average, and gradient deviation of rain rate. A maximum-likelihood algorithm, which is expressed in terms of the principal component of the previous rain-cell features, is introduced to estimate rain-cell pattern parameters from the available radar data. A detailed sensitivity analysis, which is devoted to find the best behavior in terms of root-mean-square error and correlation coefficient between the modeled and measured rain-cell features, is finally carried out.

Error Modeling Radar Rainfall Estimation Through Incorporating Rain Gauge Data Over Upper Blue Nile Basin, Ethiopia

“…The standard deviation of the total error is the component of both error terms, which is calculated by (8). Substituting the total standard deviation from both errors into probability density function produced the likelihood function as shown in (9) after some algebraic manipulation [30], [31], [32].…”

Advanced of Mathematics-Statistics Methods to Radar Calibration for Rainfall Estimation; A Review

“…The nearest neighbor fuzzy can effectively process the combined features with different data types and scales in radar target recognition [51]. In some cases the nearest neighbor used to show the similarity of radar pixel data [52]. Definitely, NN is the successful statistical technique to find appropriate result and very near the real outcomes, in radar final data.…”

Stochastic convective rain‐field simulation using a high‐resolution synoptically conditioned weather generator (HiReS‐WG)

“…The HYCELLL function generates idealized convective cell rain fields. In fact, this is the case for most of the rain cell models: based models [Capsoni et al, 1987;Montopoli and Marzano, 2007;von Hardenberg et al, 2003], and hybrid models that combine both Gaussian and exponential decay functions [Feral et al, 2003;Montopoli and Marzano, 2007]; all simulate convective cells with one rainfall peak at the cell core. In reality, rain data are much noisier.…”