Search

Machine learning algorithms are trained on a 10-yr archive of composite weather radar images in the Swiss Alps to nowcast precipitation growth and decay in the next few hours in moving coordinates (Lagrangian frame). The hypothesis of this study is that growth and decay is more predictable in mountainous regions, which represent a potential source of practical predictability by machine learning methods. In this paper, artificial neural networks (ANN) are employed to learn the complex nonlinear dependence relating the growth and decay to the input predictors, which are geographical location, mesoscale motion vectors, freezing level height, and time of the day. The average long-term growth and decay patterns are effectively reproduced by the ANN, which allows exploring their climatology for any combination of predictors. Due to the low intrinsic predictability of growth and decay, its prediction in real time is more challenging, but is substantially improved when adding persistence information to the predictors, more precisely the growth and decay and precipitation intensity in the immediate past. The improvement is considerable in mountainous regions, where, depending on flow direction, the root-mean-square error of ANN predictions can be 20%–30% lower compared with persistence. Because large uncertainty is associated with precipitation forecasting, deterministic machine learning predictions should be coupled with a model for the predictive uncertainty. Therefore, we consider a probabilistic perspective by estimating prediction intervals based on a combination of quantile decision trees and ANNs. The probabilistic framework is an attempt to address the problem of conditional bias, which often characterizes deterministic machine learning predictions obtained by error minimization. [ABSTRACT FROM AUTHOR]

Compared to traditional single-polarization radar, dual-polarization radar has a number of advantages for quantitative precipitation estimation because more information about the drop size distribution and hydrometeor type can be gleaned. In this paper, an improved dual-polarization rainfall methodology is proposed, which is driven by a region-based hydrometeor classification mechanism. The objective of this study is to incorporate the spatial coherence and self-aggregation of dual-polarization observables in hydrometeor classification and to produce robust rainfall estimates for operational applications. The S-band dual-polarization data collected from the NASA Polarimetric (NPOL) radar during the GPM Iowa Flood Studies (IFloodS) ground validation field campaign are used to demonstrate and evaluate the proposed rainfall algorithm. Results show that the improved rainfall method provides better performance than a few single- and dual-polarization algorithms in previous studies. This paper also investigates the impact of radar beam broadening on various rainfall algorithms. It is found that the radar-based rainfall products are less correlated with ground disdrometer measurements as the distance from the radar increases. [ABSTRACT FROM AUTHOR]

A prototype quantitative precipitation estimate (QPE) algorithm that utilizes specific attenuation A and specific differential phase KDP was developed for inclusion into the Multi-Radar Multi-Sensor (MRMS) system and the Weather Surveillance Radar-1988 Doppler (WSR-88D) network. Special attention is given to the optimization of the factor α used for computation of a path-integrated attenuation from a total span of differential phase along the propagation path in rain. It is suggested to estimate α from a slope of the ZDR dependence on Z in rain. The use of real-time adjusted α allows us to capture the variations of the drop size distributions, and therefore improve the QPE accuracy. It is demonstrated that the factor α is generally higher for tropical rain type compared to continental rain. Since the R(A) approach is only valid for pure rainfall, the R(KDP) relation is suggested as a complement in areas of hail contamination. The paper contains a description of the basic version of the R(A) and R(KDP) algorithm and recommendations for its further optimization. [ABSTRACT FROM AUTHOR]

This paper describes an observational study for the geometrical properties of hydrometeors during a refreezing process initiated by a winter storm in the northern Kanto Plain, Japan, on 29 January 2016, in which a subfreezing layer developed below a melting layer. The observations by using dual-polarization radar showed consistency between high values of differential reflectivity ZDR signals in midair and ice-pellet reports at the surface. The high ZDR was indicative of the sideways-oriented particles with a small axis ratio. The low ZDR signals in midair corresponded with the reports of rain or rain/ice-pellet mixtures. Observations by using a two-dimensional video disdrometer (2DVD) near the ground showed different microphysics corresponding to high ZDR and low ZDR periods. The high ZDR periods of 2DVD observations indicated that the hydrometeors exhibited dual modes of fall velocities, namely, fast-falling and slow-falling modes. The fast-falling particles were found to be deformed ice pellets with long sideways orientations that contributed to the high ZDR. The slow-falling particles were also deformed ice pellets but with a variety of orientations. This feature was rather close to that of general dry conditions of ice particles in the atmosphere. Meanwhile, the low ZDR periods of 2DVD observations indicated that the hydrometeors exhibited a single mode of fall velocity close to the terminal velocity of raindrops, but with a more spherical shape compared to raindrops. Hence, it is suggested that the high ZDR signal occurs during freezing between the initial stage of spherical ice forming and completely freezing stage of ice pellets with a variety of orientations. [ABSTRACT FROM AUTHOR]

Information about the characteristics of ice particles in clouds is necessary for improving our understanding of the states, processes, and subsequent modeling of clouds and precipitation for numerical weather prediction and climate analysis. Two NASA passive microwave radiometers, the satellite-borne Global Precipitation Measurement (GPM) Microwave Imager (GMI) and the aircraft-borne Conical Scanning Millimeter-Wave Imaging Radiometer (CoSMIR), measure vertically and horizontally polarized microwaves emitted by clouds (including precipitating particles) and Earth's surface below. In this paper, GMI (or CoSMIR) data are analyzed with CloudSat (or aircraft-borne radar) data to find polarized difference (PD) signals not affected by the surface, thereby obtaining the information on ice particles. Statistical analysis of 4 years of GMI and CloudSat data, for the first time, reveals that optically thick clouds contribute positively to GMI PD at 166 GHz over all the latitudes and their positive magnitude of 166-GHz GMI PD varies little with latitude. This result suggests that horizontally oriented ice crystals in thick clouds are common from the tropics to high latitudes, which contrasts the result of Cloud–Aerosol Lidar and Infrared Pathfinder Satellite Observations (CALIPSO) that horizontally oriented ice crystals are rare in optically thin ice clouds. [ABSTRACT FROM AUTHOR]

In a series of two papers, rain-rate retrievals based on specific attenuation A at radar X-band wavelength using the R( A) method presented by Ryzhkov et al. are thoroughly investigated. Continuous time series of overlapping measurements from two polarimetric X-band weather radars in Germany during the summers of 2011-13 are used to analyze various aspects of the method, like miscalibration correction, ground clutter contamination, partial beam blockage (PBB), sensitivity to precipitation characteristics, and sensitivity to temperature assumptions in the retrievals. In Part I of the series, the relations inherent to the R( A) method were used to calculate radar reflectivity Z from specific attenuation and it was compared with measured reflectivity to estimate PBB and calibration errors for both radars. In this paper, R( A) rain estimates are compared to R( Z) and R( KDP) retrievals using specific phase shift KDP. PBB and calibration corrections derived in Part I made the R( Z) rainfall estimates almost perfectly consistent. Accumulated over five summer months, rainfall maps showed strong effects of clutter contamination if R( KDP) is used and weaker impact on R( A). These effects could be reduced by processing the phase shift measurements with more resilience toward ground clutter contamination and by substituting problematic R( KDP) or R( A) estimates with R( Z). Hourly and daily accumulations from rain estimators are compared with rain gauge measurements; the results show that R( A) complemented by R( Z) in segments with low total differential phase shift correlates best with gauges and has the lowest bias and RMSE, followed by R( KDP) substituted with R( Z) at rain rates below 8 mm h−1. [ABSTRACT FROM AUTHOR]

In a two-part paper, radar rain-rate retrievals using specific attenuation A suggested by Ryzhkov et al. are thoroughly investigated. Continuous time series of overlapping measurements from two twin polarimetric X-band weather radars in Germany during the summers of 2011-13 are used to analyze various aspects of rain-rate retrieval, including miscalibration correction, mitigation of ground clutter contamination and partial beam blockage (PBB), sensitivity to precipitation characteristics, and the temperature assumptions of the R( A) technique. In this paper, the relations inherent to the R( A) method are used to estimate radar reflectivity Z from A and compare it to the measured Z in order to estimate PBB and calibration offsets for both radars. The fields of Z estimated from A for both radars are consistent, and the differences between Z( A) and measured Z are in good agreement with the ones calculated using either consistency relations between reflectivity at horizontal polarization Z H, differential reflectivity ZDR, and specific differential phase KDP in rain or a digital elevation model in the presence of PBB. In the analysis, the dependence of A on temperature appears to have minimal effects on the overall performance of the method. As expected, the difference between Z( A) and attenuation-corrected measured Z observations varies with rain type and exhibits a weak systematic dependency on rainfall intensity; thus, averaging over several rain events is required to obtain reliable estimates of the Z biases caused by radar miscalibration and PBB. [ABSTRACT FROM AUTHOR]

The NOAA W-band radar was deployed on a P-3 aircraft during a study of storm fronts off the U.S. West Coast in 2015 in the second CalWater (CalWater-2) field program. This paper presents an analysis of measured equivalent radar reflectivity factor Zem profiles to estimate the path-averaged precipitation rate and profiles of precipitation microphysics. Several approaches are explored using information derived from attenuation of Zem as a result of absorption and scattering by raindrops. The first approach uses the observed decrease of Zem with range below the aircraft to estimate column mean precipitation rates. A hybrid approach that combines Zem in light rain and attenuation in stronger rain performed best. The second approach estimates path-integrated attenuation (PIA) via the difference in measured and calculated normalized radar cross sections (NRCSm and NRCSc, respectively) retrieved from the ocean surface. The retrieved rain rates are compared to estimates from two other systems on the P-3: a Stepped Frequency Microwave Radiometer (SFMR) and a Wide-Swath Radar Altimeter (WSRA). The W-band radar gives reasonable values for rain rates in the range 0-10 mm h−1 with an uncertainty on the order of 1 mm h−1. Mean profiles of Zem, raindrop Doppler velocity, attenuation, and precipitation rate in bins of rain rate are also computed. A method for correcting measured profiles of Zem for attenuation to estimate profiles of nonattenuated profiles of Ze is examined. Good results are obtained by referencing the surface boundary condition to the NRCS values of PIA. Limitations of the methods are discussed. [ABSTRACT FROM AUTHOR]

Coherent power is an alternative to the conventional noise-subtracted power technique for measuring weather radar signal power. The inherent noise-canceling feature of coherent power eliminates the need for estimating and subtracting the noise component, which is required when performing conventional signal power estimation at low signal-to-noise ratio. The coherent power technique is particularly useful when averaging a high number of samples to improve sensitivity to weak signals. In such cases, the signal power is small compared to the noise power and the required accuracy of the estimated noise power may be difficult to achieve. This paper compares conventional signal power estimation with the coherent power measurement technique by investigating bias, standard deviation, and probability of false alarm and detection rates as a function of signal-to-noise ratio and threshold level. This comparison is performed using analytical expressions, numerical simulations, and analysis of cloud and precipitation data collected with the airborne solid-state Ka-band precipitation radar (KPR) operated by the University of Wyoming. [ABSTRACT FROM AUTHOR]

Water vapor has a strong influence on the evolution of heavy precipitation events due to the huge latent heat associated with the phase change process of water. Accurate monitoring of atmospheric water vapor distribution is thus essential in predicting the severity and life cycle of heavy rain. This paper presents a systematic study on the application of tomographic solutions to investigate water vapor variations during heavy precipitation events. Using global positioning system (GPS) observations, the wet refractivity field was constructed at a temporal resolution of 30 min for three heavy precipitation events occurring in Hong Kong, China, in 2010-14. The zenith wet delay (ZWD) is shown to be a good indicator in observing the water vapor evolution in heavy rain events. The variabilities of water vapor at five altitude layers (<1000, 1000-2000, 2000-3000, 3000-5000, and >5000 m) were examined. It revealed that water vapor above 3000 m has larger fluctuation than that under 3000 m, though it accounts for only 10%-25% of the total amount of water vapor. The relative humidity fields derived from tomographic results revealed moisture variation, accumulation, saturation, and condensation during the heavy rain events. The water vapor variabilities observed by tomography have been validated using European Centre for Medium-Range Weather Forecasts (ECMWF) reanalysis and radiosonde data. The results positively demonstrated the potential of using water vapor tomographic technique for detecting and monitoring the evolution of heavy rain events. [ABSTRACT FROM AUTHOR]

The U.S. Weather Surveillance Radar-1988 Doppler (WSR-88D) network has provided meteorologists and hydrologists with quantitative precipitation observations at an unprecedented high spatial-temporal resolution since its deployment in the mid-1990s. Since each single radar can only cover a maximum range of 460 km, a mosaic of multiple-radar observations is needed to generate any national-scale products. The Multi-Radar Multi-Sensor (MRMS) system utilizes a physically based two-dimensional mosaicking algorithm of the WSR-88D data to generate seamless national quantitative precipitation estimation (QPE) products. For areas covered by multiple radars, the mosaicking scheme first determines if precipitation is present by checking the lowest-altitude observation. If the lowest observed radar data indicate no precipitation, then the mosaicked value is set to no precipitation. Otherwise, a weighted mean of multiple-radar observations is taken as the mosaicked value. The weighting function is based on multiple factors, including the distance from the radar and the height of the observation with respect to the melting layer. The mosaic algorithm uses the physically lowest radar observations with no/little blockage while maintaining a spatial continuity in the mosaicked field. The performance of the MRMS seamless radar mosaic algorithm was examined for various precipitation events of different characteristics. The results of these case evaluations are presented in this paper. [ABSTRACT FROM AUTHOR]

This paper presents a novel, simple method to correct reflectivity measurements of weather radars that operate in attenuation-influenced frequency bands using observations from less attenuated radar systems. In recent years radar systems operating in the X-band frequency range have been developed to provide precipitation fields for areas of special interest in high temporal (≤1 min) and spatial (≤250 m) resolution in complement to nationwide radar networks. However, X-band radars are highly influenced by attenuation. C- and S-band radars typically have coarser resolution (250 m-1 km and 5 min) but are less affected by attenuation. Correcting for attenuation effects in simple (non-Doppler) single-polarized X-band radars remains challenging and is often dependent on restriction parameters, for example, those derived from mountain returns. Therefore, these algorithms are applicable only in limited areas. The method proposed here uses measurements from C-band radars and hence can be applied in all regions covered by nationwide C- (or S-) band radar networks. First, a single scan of X-band radar measurements is used exemplary to identify advantages and disadvantages of the novel algorithm compared to a standard single radar algorithm. The performance of the correction algorithms in different types of precipitation is examined in nine case studies. The proposed method provides very promising results for each type of precipitation. Additionally, it is evaluated in a 5-month comparison with Micro Rain Radar (MRR) observations. The bias between uncorrected X-band radar and MRR data is nearly eliminated by the attenuation correction algorithm, and the RMSE is reduced by 20% while the correlation of ~0.9 between both systems remains nearly constant. [ABSTRACT FROM AUTHOR]

In this paper, the operational Global Precipitation Measurement (GPM) mission combined radar-radiometer algorithm is thoroughly described. The operational combined algorithm is designed to reduce uncertainties in GPM Core Observatory precipitation estimates by effectively integrating complementary information from the GPM Dual-Frequency Precipitation Radar (DPR) and the GPM Microwave Imager (GMI) into an optimal, physically consistent precipitation product. Although similar in many respects to previously developed combined algorithms, the GPM combined algorithm has several unique features that are specifically designed to meet the GPM objectives of deriving, based on GPM Core Observatory information, accurate and physically consistent precipitation estimates from multiple spaceborne instruments, and ancillary environmental data from reanalyses. The algorithm features an optimal estimation framework based on a statistical formulation of the Gauss-Newton method, a parameterization for the nonuniform distribution of precipitation within the radar fields of view, a methodology to detect and account for multiple scattering in Ka-band DPR observations, and a statistical deconvolution technique that allows for an efficient sequential incorporation of radiometer information into DPR precipitation retrievals. [ABSTRACT FROM AUTHOR]

This paper demonstrates the impact of the enhancement in detectability by the dual-frequency precipitation radar (DPR) on board the Global Precipitation Measurement (GPM) core observatory. By setting two minimum detectable reflectivities-12 and 18 dB Z-artificially to 6 months of GPM DPR measurements, the precipitation occurrence and volume increase by ~21.1% and ~1.9%, respectively, between 40°S and 40°N. GPM DPR is found to be able to detect light precipitation, which mainly consists of two distinct types. One type is shallow precipitation, which is most significant for convective precipitation over eastern parts of subtropical oceans, where deep convection is typically suppressed. The other type is probably associated with lower parts of anvil clouds associated with organized precipitation systems. While these echoes have lower reflectivities than the official value of the minimum detectable reflectivity, they are found to mostly consist of true precipitation signals, suggesting that the official value may be too conservative for some sort of meteorological analyses. These results are expected to further the understanding of both global energy and water budgets and the diabatic heating distribution. [ABSTRACT FROM AUTHOR]

A heavy rainfall event over a 2-h period on 8 July 2013 caused significant flash flooding in the city of Toronto and produced 126 mm of rain accumulation at a gauge located near the Toronto Pearson International Airport. This paper evaluates the quantitative precipitation estimates from the nearby King City C-band dual-polarized radar (WKR). Horizontal reflectivity Z and differential reflectivity ZDR were corrected for attenuation using a modified ZPHI rain profiling algorithm, and rain rates R were calculated from R( Z) and R( Z, ZDR) algorithms. Specific differential phase KDP was used to compute rain rates from three R( KDP) algorithms, one modified to use positive and negative KDP, and an R( KDP, ZDR) algorithm. Additionally, specific attenuation at horizontal polarization A was used to calculate rates from the R( A) algorithm. High-temporal-resolution rain gauge data at 44 locations measured the surface rainfall every 5 min and produced total rainfall accumulations over the affected area. The nearby NEXRAD S-band dual-polarized radar at Buffalo, New York, provided rain-rate and storm accumulation estimates from R( Z) and S-band dual-polarimetric algorithm. These two datasets were used as references to evaluate the C-band estimates. Significant radome attenuation at WKR overshadowed the attenuation correction techniques and resulted in poor rainfall estimates from the R( Z) and R( Z, ZDR) algorithms. Rainfall estimation from the Brandes et al. R( KDP) and R( A) algorithms were superior to the other methods, and the derived storm total accumulation gave biases of 2.1 and −6.1 mm, respectively, with correlations of 0.94. The C-band estimates from the Brandes et al. R( KDP) and R( A) algorithms were comparable to the NEXRAD S-band estimates. [ABSTRACT FROM AUTHOR]

Dual-polarization Doppler radar has proven useful for the estimation of hydrometeor content and the classification of hydrometeor type. Recent studies have leveraged dual-polarization-specific information to produce improved assimilated cloud and precipitation fields from the warm rain (above freezing) portion of deep convective storms. While the strengths of dual-polarization radar observations have been conclusively shown for rain and hail hydrometeors, it is less clear how much information is provided in mixed-phase and ice-only regions. In this paper, a Markov chain Monte Carlo (MCMC) algorithm is used to examine the information content of dual-polarization-specific variables in the ice-phase region of a convective storm. Results are used to quantify how much information is added by specific differential phase and radar correlation coefficient, as well as how this information is degraded when the assumed particle size distribution and particle density are allowed to vary. It is found that dual-polarization-specific observations ( Kdp and ρhv) provide significant information on rimed ice content, and moderate information on pristine ice, especially where snow mass is more than 10% of the total volume hydrometeor mass. There is a significant reduction in information content for rain and a near-complete loss of information for graupel-hail and snow when the particle size distribution and ice particle densities are not well known, and there are systematic changes in radar information gain and loss with changes in hydrometeor mass. The results highlight the need for a thorough exploration of forward model sensitivities prior to performing radar data assimilation. [ABSTRACT FROM AUTHOR]

The Tropical Rainfall Measuring Mission (TRMM) Precipitation Radar (PR) measures reflectivity downward from space and provides observations of the vertical distributions of precipitation over land as well as the ocean. It overpasses the southern part of the Korean Peninsula where (i) a dense network of operational S-band scanning radars is available and (ii) various types of precipitation occur. By utilizing a 3D reflectivity composite from the ground S-band radar (GR) observations, this paper shows a comparison of reflectivity profiles observed with both PR and GR focusing on their vertical structure. For four cases of widespread rain, visual and statistical analyses show that PR attenuation-corrected reflectivity agrees closely with reflectivity observed from the GR composite below the melting layer. Above and within the melting layer, PR is affected critically by its sensitivity while GR beam broadening at far ranges causes systematic differences in the PR-GR comparisons. For four cases of convective rain, PR underestimates the mean reflectivities by 1-3 dB compared with those from GR at low levels where precipitation attenuation is significant toward the ground. In these cases, the low sensitivity of PR results in a small number of matched points for weak echoes. Also, the PR-GR discrepancy for the convective case is more affected by time mismatching. [ABSTRACT FROM AUTHOR]

Although satellite precipitation estimates provide valuable information for weather and flood forecasts, infrared (IR) brightness temperature (BT)-based algorithms often produce large errors for precipitation detection and estimation during deep convective systems (DCSs). As DCSs produce greatly varying precipitation rates below similar IR BT retrievals, using IR BTs alone to estimate precipitation in DCSs is problematic. Classifying a DCS into convective-core (CC), stratiform (SR), and anvil cloud (AC) regions allows an evaluation of estimated precipitation distributions among DCS components to supplement typical quantitative precipitation estimate (QPE) evaluations and to diagnose these IR-based algorithm biases. This paper assesses the performance of the National Mosaic and Multi-Sensor Next Generation Quantitative Precipitation Estimation System (NMQ Q2), and a simplified version of the Self-Calibrating Multivariate Precipitation Retrieval (SCaMPR) algorithm, over the state of Oklahoma using Oklahoma Mesonet observations. While average annual Q2 precipitation estimates were about 35% higher than Mesonet observations, strong correlations exist between these two datasets for multiple temporal and spatial scales. Additionally, the Q2-estimated precipitation distribution among DCS components strongly resembled the Mesonet-observed distribution, indicating Q2 can accurately capture the precipitation characteristics of DCSs despite its wet bias. SCaMPR retrievals were typically 3-4 times higher than Mesonet observations, with relatively weak correlations during 2012. Overestimates from SCaMPR retrievals were primarily caused by precipitation retrievals from the anvil regions of DCSs when collocated Mesonet stations recorded no precipitation. A modified SCaMPR retrieval algorithm, employing both cloud optical depth and IR temperature, has the potential to make significant improvements to reduce the wet bias of SCaMPR retrievals over anvil regions of a DCS. [ABSTRACT FROM AUTHOR]