Traditional radar-based rainfall estimation is typically done by known functional relationships between the rainfall intensity (R) and radar measurables, such as R–Z_h, R–(Z_h, Z_DR), etc. One of the biggest advantages of machine learning algorithms is the applicability to a non-linear relationship between a dependent variable and independent variables without any predefined relationships. We explored the potential use of two supervised machine learning methods (regression tree and random forest) in rainfall estimation using dual-polarization radar variables. The regression tree does not require normalization and scaling of data; however, this method is quite unstable since each split depends on the parent split. Since the random forest is an ensemble method of regression trees, it has less variability in prediction compared with regression trees, but consumes more computer resources. We considered several different configurations for machine learning algorithms with different sets of dependent and independent variables. The random forest model was appropriately tuned. In the test of variable importance, the specific differential phase (differential reflectivity) was the most important variable to predict the rainfall rate (residual that is the difference between the true rainfall rate and the one estimated from the R–Z relationship). The models were evaluated by 10-fold cross-validation. The best model was the random forest model using a residual with the non-classified training set. The results indicated that the machine learning algorithms outperformed the traditional R–Z relationship. Then, we applied the best machine learning model to an S-band dual-polarization radar (Mt. Myeonbong) and validated the result with ground rain gauges. The results of the application to radar data showed that the estimates of the residuals had spatial variability. The stratiform and weak rain areas had positive residuals while convective areas had negative residuals, indicating that the spatial error structure driven by the R–Z relationship was well captured by the model. The rainfall rates of all pixels over the study area were adjusted with the estimated residuals. The rainfall rates adjusted by residual showed excellent agreement with the rain gauge, especially at high rainfall rates.

An accurate classification of the precipitation type is important for forecasters, particularly in the winter season. We explored the capability of three supervised machine learning (ML) methods (decision tree, random forest, and support vector machine) to determine ground precipitation types (no precipitation, rain, mixed, and snow) for winter precipitation. We provided information on the particle characteristics within a radar sampling volume and the environmental condition to the ML model with the simultaneous use of polarimetric radar variables and thermodynamic variables. The ML algorithms were optimized using predictor selection and hyperparameter tuning in order to maximize the computational efficiency and accuracy. The random forest (RF) had the highest skill scores in all precipitation types and outperformed the operational scheme. The spatial distribution of the precipitation type from the RF model showed a good agreement with the surface observation. As a result, RF is recommended for the real-time precipitation type classification due to its easy implementation, computational efficiency, and satisfactory accuracy. In addition to the validation, this study confirmed the strong dependence of precipitation type on wet-bulb temperature and a 1000–850 hPa layer thickness. The results also suggested that the base heights of the radar echo are useful in discriminating non-precipitating area.

Retrieving raindrop size distribution (DSD) is essential to understanding precipitation processes. Conventional approaches based on polarimetric radar (e.g., polynomial regression) struggle to accurately capture the inherent nonlinearity between DSD parameters and radar measurables. In contrast, machine learning (ML) algorithms offer a promising solution as it effectively models the complex non‐linear relationship. We have developed an ML algorithm to retrieve DSD parameters using polarimetric radar variables in a framework of double‐moment normalization. The potentially stable and invariant double‐moment normalized DSD enables the applicability of the algorithm in any climatic regime or any precipitation system. To improve the robustness of the model to measurement noises, we employed training samples with random noise. All ML algorithms outperformed the conventional method, with the random forest being the best model. This study highlights the effectiveness of the developed algorithm as a tool for understanding the DSD characteristics from polarimetric radar measurements.

Estimating precipitation area is important for weather forecasting as well as real-time application. This paper aims to develop an analytical framework for efficient precipitation area estimation using S-band dual-polarization radar measurements. Several types of factors, such as types of sensors, thresholds, and models, are considered and compared to form a data set. After building the appropriate data set, this paper yields a rigorous comparison of classification methods in statistical (logistic regression and linear discriminant analysis) and machine learning (decision tree, support vector machine, and random forest). To achieve better performance, spatial classification is considered by incorporating latitude and longitude of observation location into classification, compared with non-spatial classification. The data used in this study were collected by rain detector and present weather sensor in a network of automated weather systems (AWS), and an S-band dual-polarimetric weather radar during ten different rainfall events of varying lengths. The mean squared prediction error (MSPE) from leave-one-out cross validation (LOOCV) is computed to assess the performance of the methods. Of the methods, the decision tree and random forest methods result in the lowest MSPE, and spatial classification outperforms non-spatial classification. Particularly, machine-learning-based spatial classification methods accurately estimate the precipitation area in the northern areas of the study region.

The accurate prediction of lightning is crucial for forecasters to respond effectively to its related hazards. The rapid development and confined spatial extent of convective storms, in which lightning frequently occurs, pose considerable challenges for accurately predicting their locations using numerical weather prediction (NWP) models. Lightning occurrence is often prognosed using thermodynamic parameters, convective available potential energy (CAPE), the severe weather threat index (SWEAT), the lifted index (LI), etc. A high-resolution NWP model provides a prediction of these thermodynamic parameters at high spatiotemporal resolution with high accuracy for a few hours. However, a complicated algorithm is required to handle all the useful high-resolution variables from the NWP model. The recently emerging machine learning technique can solve this issue by properly handling these “big data” without any model distributional assumption. In this study, we developed a random forest algorithm for nowcasting and very short-range forecasting (useful for ~6 h), named LightningRF. LightningRF was trained by using lightning occurrence as a response variable and characteristic parameters from the NWP as predictors. It was also applied to analysis and forecast fields, showing a high probability of lightning within the observed lightning regions. This highlights the potential of helping forecasters improve their lightning forecasting skills using real-time probabilistic forecasts from a trained model.

The probability density function of drops is difficult to model. Current approaches make assumptions that are often problematic, as they allow negative values for the mean of the distribution. While the statistical goodness of fit of those models might be reasonable for precipitation radar estimation, the situation is unsatisfactory if a fully consistent physical modeling of precipitation across scales is desired. This is the case of weather and climate models. This paper discusses a model that satisfies mathematical and physical consistency. The model can be seamlessly integrated into the parameterizations of the microphysics of precipitation and is tested on an extensive disdrometer data set. Comparison with existing models shows that the new method has substantial practical and theoretical advantages. The research has implications in elucidating the role of clouds in the climate sensitivity of climate models.