Numerical weather models such as WRF (Weather Research and Forecasting) are increasingly used in studies on water resources. However, they have suffered from relatively poor performance in rainfall estimation. Among the various influential factors, a critical parameter in the WRF model rainfall retrieval is raindrop size distribution (DSD), which has not been fully explored. The analysis of sensitivity and uncertainty of the DSD model accuracy is significant for rainfall forecasts based on mesoscale numerical weather prediction (NWP) models. A WRF-disdrometer integrated error assessment framework is developed to analyze the accuracy and sensitivity of DSD parameterizations of gamma distribution in WRF rainfall simulation. This study adopts three different microphysics parameterizations (Morrison, WDM6, and Thompson aerosol-aware) to simulate the DSD of approximately one hundred rainfall events in Chilbolton, UK that are categorized into 12 scenarios based on the season, rainfall evenness, and rainfall rate. The Thompson aerosol-aware microphysics scheme shows the best performance among the three. In comparisons of WRF rainfall simulations across different scenarios of evenness and rainfall rate, a higher accuracy is obtained with more even rain and a higher rainfall rate. The sensitivity results of different DSD parameterizations indicate that the sensitivity to the intercept parameter 0 is pronouncedly higher than those to the shape parameter μ and slope parameter λ for all studied schemes. The overall WRF rainfall shows a trend of slight underestimation followed by overestimation as μ increases; further, the rainfall is overestimated when 10 0 or λ decreases and is underestimated when it increases and then remains constant. Comparisons of different scenarios reveal that variations of DSD parameters of even rain have a relatively high impact on rainfall recognizability, and the DSD parameterizations show a higher sensitivity for rainfall with a low rate. Moreover, the sensitivity discrimination is not clear among the rainfall of different seasons. The uncertainty assessment of the WRF rainfall retrieval caused by the shape parameter suggests that a gamma DSD model with a variable shape parameter should be developed according to the evenness, rainfall rate, and microphysics parameterizations by using the WRF model. Some modified algorithms of the WRF gamma DSD model for achieving better accuracy in WRF rainfall retrievals will be explored in future studies with various climatic regimes by adjusting the DSD parameterization based on the assimilation of measured data.

Radar‐gauge rainfall discrepancies are considered to originate from radar rainfall measurements while ignoring the fact that radar observes rain aloft while a rain gauge measures rainfall on the ground. Observations of raindrops observed aloft by weather radars consider that raindrops fall vertically to the ground without changing in size. This premise obviously does not stand because raindrop location changes due to wind drift and raindrop size changes due to evaporation. However, both effects are usually ignored. This study proposes a fully formulated scheme to numerically simulate both raindrop drift and evaporation in the air and reduces the uncertainties of radar rainfall estimation. The Weather Research and Forecasting model is used to simulate high‐resolution three‐dimensional atmospheric fields. A dual‐polarization radar retrieves the raindrop size distribution for each radar pixel. Three schemes are designed and implemented using the Hameldon Hill radar in Lancashire, England. The first considers only raindrop drift, the second considers only evaporation, and the last considers both aspects. Results show that wind advection can cause a large drift for small raindrops. Considerable loss of rainfall is observed due to raindrop evaporation. Overall, the three schemes improve the radar‐gauge correlation by 3.2%, 2.9%, and 3.8% and reduce their discrepancy by 17.9%, 8.6%, and 21.7%, respectively, over eight selected events. This study contributes to the improvement of quantitative precipitation estimation from radar polarimetry and allows a better understanding of precipitation processes.

The understanding of large-scale rainfall microphysical characteristics plays a significant role in meteorology, hydrology and natural hazards managements. Traditional instruments for estimating raindrop size distribution (DSD), including disdrometers and ground dual-polarization radars, are available only in limited areas.However, the development of space-based radars and mesoscale numerical weather prediction models would allow for DSD estimation on a large scale. This study investigated the performance of the weather research and forecasting (WRF) model and the global precipitation measurement mission (GPM) dual-frequency precipitation radar for DSD retrieval under different conditions. The DSD parameters (D m and N w ), rain rate (R), rainfall kinetic energy (KE) and radar reflectivity (Z) were estimated in Chilbolton, United Kingdom, by using long-term disdrometer observations for validation. The rainfall kinetic energy-rain rate (KE-R) and radar reflectivity-rain rate (Z-R) relationships were explored using a disdrometer, the WRF model and GPM. It was found that the DSD parameter distribution trends of the three approaches are similar although the WRF model has larger D m and smaller N w values. In terms of the rainfall microphysical relationship, GPM performs better when both Ku-and Kaband precipitation radars (KuPR and KaPR) observe precipitation simultaneously (R > 0.5 mm h À1 ), while the WRF model shows high accuracy in light rain (R < 0.5 mm h À1 ). The fusion of GPM and WRF model is recommended for the improved understanding of rainfall microphysical characteristics in ungauged areas.

In modeling the radar rainfall uncertainty, rain gauge measurement is generally regarded as the areal “true” rainfall. However, the inconsistent scales between radar and gauge may introduce a new uncertainty (also known as gauge representative uncertainty), which is erroneously identified as radar rainfall uncertainty and therefore called pseudouncertainty. It is crucial to comprehend what percentage of the estimated radar rainfall uncertainty actually stems from such pseudouncertainty rather than radar rainfall itself. For this reason, based on a fully formulated radar rainfall uncertainty model, this study aims to explore how the gauge representative error affects the distribution, spatial dependence, and temporal dependence of hourly accumulated radar rainfall uncertainty, and consequently affects the produced radar rainfall uncertainty band. Three group scenarios that delineate various degrees of gauge representative errors were designed to configure and run the uncertainty model. In the setting of a long-term analysis (almost 7 years) of the Brue catchment in the United Kingdom, we found that the gauge representative error affected the simulation of the marginal distribution of radar rainfall error, and had a considerable effect on temporal dependence estimation of radar rainfall uncertainty. The spread of the rainfall uncertainty band decreased with the growth of the gauge density in a radar pixel. The scenario with the lowest representative error only had 78% uncertainty spread of the scenario that has the largest error. This indicated there was a large impact of the representative error on radar rainfall uncertainty models. It is hoped that more catchments with diverse climate and geographical conditions and more radar data with various spatial scales could be explored by the research community to further investigate this crucial issue.

No abstract

No abstract

Rainfall estimation using the weather research and forecasting (WRF) model is sensitive to physical parameterizations and downscaling configurations. Concerned with the correlations between physical parameterizations and dynamical downscaling, these significant issues were considered simultaneously in this study, and WRF‐based ensembles were integrated and used to estimate eight representative rainfall events. The results revealed that both rainfall estimates and intensities were sensitive to downscaling configurations. Specifically, light rainfall events that were homogeneous over space and time were estimated well using a 5 km horizontal resolution and were overestimated with a 1 km resolution where the planetary boundary layer (PBL) schemes were probably the source of positive errors in light rain conditions. However, when the rainfall was intense and displayed spatiotemporal heterogeneity, the rainfall peaks and volume were only well estimated with a 1 km resolution where cumulus (CU) schemes were the dominant schemes, demonstrating that higher resolutions could better reproduce rainfall patterns for wetter cases. At a 10 km horizontal resolution, the simulations did not display much accuracy, and different physics schemes did not make a substantial difference. Therefore, the rainfall characteristics cannot be ignored. More importantly, the contributions of microphysics, CU and PBL ensembles were quantified, and the sensitivities of the rainfall estimates to three downscaling configurations were studied.

High-resolution radar rainfall data have great potential for rainfall predictions up to 6 h ahead (nowcasting); however, conventional extrapolation approaches based on in-built physical assumptions yield poor performance at longer lead times (3–6 h), which limits their operational utility. Moreover, atmospheric factors in radar estimate errors are often ignored. This study proposed a radar rainfall nowcasting method that attempts to achieve accurate nowcasting of 6 h using long short-term memory (LSTM) networks. Atmospheric conditions were considered to reduce radar estimate errors. To build radar nowcasting models based on LSTM networks (LSTM-RN), approximately 11 years of radar, gauge rainfall, and atmospheric data from the UK were obtained. Compared with the models built on optical flow (OF-RN) and random forest (RF-RN), LSTM-RN had the lowest root-mean-square errors (RMSE), highest correlation coefficients (COR), and mean bias errors closest to 0. Furthermore, LSTM-RN showed a growing advantage at longer lead times, with the RMSE decreasing by 17.99% and 7.17% compared with that of OF-RN and RF-RN, respectively. The results also revealed a strong relationship between LSTM-RN performance and weather conditions. This study provides an effective solution for nowcasting radar rainfall at long lead times, which enhances the forecast value and supports practical utility.