In a recent BAMS article, it is argued that community-based Open Source Software (OSS) could foster scientific progress in weather radar research, and make weather radar software more affordable, flexible, transparent, sustainable, and interoperable.Nevertheless, it can be challenging for potential developers and users to realize these benefits: tools are often cumbersome to install; different operating systems may have particular issues, or may not be supported at all; and many tools have steep learning curves.To overcome some of these barriers, we present an open, community-based virtual machine (VM). This VM can be run on any operating system, and guarantees reproducibility of results across platforms. It contains a suite of independent OSS weather radar tools (BALTRAD, Py-ART, wradlib, RSL, and Radx), and a scientific Python stack. Furthermore, it features a suite of recipes that work out of the box and provide guidance on how to use the different OSS tools alone and together. The code to build the VM from source is hosted on GitHub, which allows the VM to grow with its community.We argue that the VM presents another step toward Open (Weather Radar) Science. It can be used as a quick way to get started, for teaching, or for benchmarking and combining different tools. It can foster the idea of reproducible research in scientific publishing. Being scalable and extendable, it might even allow for real-time data processing.We expect the VM to catalyze progress toward interoperability, and to lower the barrier for new users and developers, thus extending the weather radar community and user base.

Weather radar analysis has become increasingly sophisticated over the past 50 years, and efforts to keep software up to date have generally lagged behind the needs of the users. We argue that progress has been impeded by the fact that software has not been developed and shared as a community. Recently, the situation has been changing. In this paper, the developers of a number of open-source software (OSS) projects highlight the potential of OSS to advance radar-related research. We argue that the community-based development of OSS holds the potential to reduce duplication of efforts and to create transparency in implemented algorithms while improving the quality and scope of the software. We also conclude that there is sufficiently mature technology to support collaboration across different software projects. This could allow for consolidation toward a set of interoperable software platforms, each designed to accommodate very specific user requirements.

The estimation of surface rainfall from reflectivity data derived from weather radar has been much studied over many years. It is now clear that central to this problem is the adjustment of these data for the impacts of vertical variations in the reflectivity. In this paper a new procedure (known as Down-to-Earth, DTE) is proposed and tested for combining radar measurements aloft with information from a numerical weather-prediction (NWP) model and an analysis system. The procedure involves the exploitation of moist cloud physics in an attempt to account for physical processes impacting on precipitation during its descent from the height of radar echo measurements to the surface. The application of DTE leads to increased underestimation in the radar measurements compared to precipitation gauge observations at short and intermediate radar ranges (0-120 km), but is successful at reducing the bias at further ranges. However the application of DTE does not lead to significant decreases in the random error of the surface rain rate estimate. No improvement is made when attempting to account for the precipitation phase measured by radar. It is concluded that further work on radar data quality control, along with improvements to the NWP model, are essential to improve upon results using such a physically based procedure.

Atmospheric state variables from a Numerical Weather Prediction (NWP) model are combined with analyzed cloud base heights in a neural network, with the objective to model corresponding cloud water profiles. It was found that the neural network was incapable of resolving the inherently non-linear vertical cloud water distributions. Copyright (C) 2004 Royal Meteorological Society

A Doppler radar wind data assimilation system has been developed for the three-dimensional variational data assimilation (3DVAR) scheme of the High Resolution Limited Area Model (HIRLAM). Radar wind observations can be input for the multivariate HIRLAM 3DVAR either as radial wind superobservations (SOs) or as vertical profiles of horizontal wind obtained with the velocity-azimuth display (VAD) technique. The radar wind data handling system, including data processing, quality control, and observation operators for the 3DVAR, are described and evaluated. Background error standard deviation (sigma(b)) in observation space for wind and radial wind have been estimated by the so-called randomization method. The derived values of sigma(b) are used in the quality control of observations and also in the assignment of radar wind observation error standard deviations (sigma(o)). Parallel data assimilation and forecast experiments confirm reasonably tuned error statistics and indicate a small positive impact of radar wind data on the verification scores, for both inputs.

A simple and pragmatic method utilising the difference between analysed near-surface and Meteosat IR temperatures (DeltaT) is presented and applied with the aim of identifying and removing non-precipitation echoes in weather radar composite imagery. Despite inherent deficiencies in these multisource data, such as lower spatial and temporal resolutions relative to the radar data, DeltaT is demonstrated to efficiently identify efficiently those areas void of potentially precipitating clouds, and to remove radar echoes in them. A set of 243 manually analysed composites from the summer of 2000 was used to evaluate the method. False alarm rates (FAR), percent correct (PC) and Hanssen-Kuipers skill (HKS) scores were calculated from standard contingency tables for five echo classes: weak, strong, land, sea, and all. FAR was lowered in all classes, PC was generally raised by a few percent to be over 95%, while HKS either remained unchanged or was slightly lowered through the application of DeltaT. These results indicate that DeltaT successfully removes a significant amount of non-precipitation, sometimes at the expense of a small amount of true precipitation. This penalty is larger over sea, which indicates that the method may need to be tuned differently for land and sea environments. This method may act as a foundation on which improvements to radar data quality control can be made with the introduction Of new and improved satellite instrumentation such as that found on board the Meteosat Second Generation platform. However, this type of method should remain complementary to improved signal processing and radar data analysis techniques.

Precipitation gauge measurements suffer from several sources Of error, the most significant of which is the wind error caused by the flow distortion about the gauge orifice. An existing statistical Dynamic Correction Model (DCM) has been implemented with the intent to perform a systematic correction of precipitation measurements front gauges found in and near the Baltic Sea's drainage basin. The DCM implementation makes use of hourly gridded meteorological variables from an operational mesoscale analysis system. precipitation amounts are disaggregated into hourly components, corrected, and then summed back to yield corrected 12-hour accumulations. Sensitivity Studies for shielded H & H-90, Tretyakov, SMHI, and unshielded Hellmann gauge types demonstrate the behaviour of the DCM; the H & H-90 gauge requires the least amount of correction whereas the unshielded Hellmann gauge requires by far the most. This DCM implementation has been evaluated using two years of independent gauge data from the so-called Double Fence Intercomparison Reference (DFIR) gauge, along with independent H & H-90 observations, at Jokioinen, Finland. The results show that the H & H-90 gauge underestimates precipitation by around 8% on average and that the implementation appears to yield results which are fully consistent with previous findings and experience at this site. A second evaluation was performed with one year of measurements from Kiel, Germany, using data from a ship rain gauge (SRG) as a reference and data from two Hellmann gauges. one co-located with the SRG and the other 5.6 km distant. The results from this evaluation are more ambiguous but reveal both an overcorrection and an increased variability in the derived relation compared with uncorrected observations, one explanation being a well shielded site which the method, by its general nature, does not take into account. Although uncertainties remain in the treatment of measurements from some gauge types, systematic correction using this DCM should lead to more accurate measurements for use in hydrometeorological applications. (C) 2004 Elsevier B.V. All rights reserved.

This paper describes the evaluation of a-combined radar and passive microwave dataset obtained during the PIDCAP study of the Baltic Sea Experiment (BALTEX), where three-dimensional volumes of data from the Gotland radar were obtained timed according to the overpasses of the DMSP-satellites F10 and F13. Both satellites are 'equipped with a Special Sensor Microwave/Imager (SSM/I), suitable for precipitation retrievals. We compare radar precipitation estimates, convolved to the native resolution of the SSM/I, at different altitudes with polarization and scattering indices (S-85) derived from the SSM/I. For all 22 overpasses investigated here radar precipitation estimates at 3-4 km altitude correlate well with the SSM/I-derived S-85 (average correlation coefficient = 0.70). Although more directly linked to surface precipitation, polarization indices have been found to be less correlated with radar data, due to limitations inherent in the remote sensing of precipitation at higher latitudes. A stratification of the dataset into frontal and convective events revealed significant variations in these relationships for different types of precipitation events, thus reflecting different cloud microphysical processes associated with precipitation initialization. The relationship between S85 and radar rain estimates at higher altitudes varies considerably for different convective and frontal events. The sensitivity of S-85 to radar-derived rain rate ranges from 3.1 K mm(-1) h(-1) for a strong convective event to about 25 K mm(-1) h(-1) for the frontal and about 70 mm(-1) h(-1) for the small-scale convective events. For extrapolated surface precipitation estimates, sensitivities decrease to 14 mm(-1) h(-1) and 25 mm(-1) h(-1) for frontal and small-scale convective precipitation, respectively.