This study investigates the statistical characteristics of raindrop size distributions (DSDs) in monsoon season with observations collected by the second-generation Particle Size and Velocity (Parsivel²) disdrometer located in Zhuhai, southern China. The characteristics are quantified based on convective and stratiform precipitation classified using the rainfall intensity and total number of drops. On average of the whole dataset, the DSD characteristic in southern China consists of a higher number concentration of relatively small-sized drops when compare with eastern China and northern China, respectively. In the meanwhile, the D_m and log₁₀N_w scatter plots prove that the convective rain in monsoon season can be identified as maritime-like cluster. The DSD is in good agreement with a three-parameter gamma distribution, especially for the medium to large raindrop size. Using filtered data observed by Parsivel² disdrometer, a new Z–R relationship, Z = 498R^1.3, is derived for convective rain in monsoon season in southern China. These results offer insights of the microphysical nature of precipitation in Zhuhai during monsoon season, and provide essential information that may be useful for precipitation retrievals based on weather radar observations.

The China Meteorological Administration has deployed the China New Generation Weather Radar (CINRAD) network for severe weather detection and to improve initial conditions for numerical weather prediction models. The CINRAD network consists of 217 radars comprising 123 S-band and 94 C-band radars over mainland China. In this paper, a high-resolution digital elevation model (DEM) and beam propagation simulations are used to compute radar beam blockage and evaluate the effective radar coverage over China. Results show that the radar coverage at a height of 1 km above ground level (AGL) is restricted in complex terrain regions. The effective coverage maps at heights of 2 km and 3 km AGL indicate that the Yangtze River Delta, the Pearl River Delta, and North China Plain have more overlapping radar coverage than other regions in China. Over eastern China, almost all areas can be sampled by more than 2 radars within 5 km above mean sea level (MSL), but the radars operating in Qinghai-Tibet Plateau still suffer from serious beam blockage caused by intervening terrain. Overall, the radars installed in western China suffer from much more severe beam blockage than those deployed in eastern China. Maps generated in this study will inform users of the CINRAD data of their limitations for use in precipitation estimation, as inputs to other weather and hydrological models, and for satellite validation studies.

This study assesses the performance of the latest version 05B (V5B) Integrated Multi-Satellite Retrievals for Global Precipitation Measurement (GPM) (IMERG) Early and Final Runs over southern China during six extremely heavy precipitation events brought by six powerful typhoons from 2016 to 2017. Observations from a dense network composed of 2449 rain gauges are used as reference to quantify the performance in terms of spatiotemporal variability, probability distribution of precipitation rates, contingency scores, and bias analysis. The results show that: (1) both IMERG with gauge calibration (IMERG_Cal) and without gauge correction (IMERG_Uncal) generally capture the spatial patterns of storm-accumulated precipitation with moderate to high correlation coefficients (CCs) of 0.57–0.87, and relative bias (RB) varying from −17.21% to 30.58%; (2) IMERG_Uncal and IMERG_Cal capture well the area-average hourly series of precipitation over rainfall centers with high CCs ranging from 0.78 to 0.94; (3) IMERG_Cal tends to underestimate precipitation especially the rainfall over the rainfall centers when compared to IMERG_Uncal. The IMERG Final Run shows promising potentials in typhoon-related extreme precipitation storm applications. This study is expected to give useful feedbacks about the latest V5B Final Run IMERG product to both algorithm developers and the scientific end users, providing a better understanding of how well the V5B IMERG products capture the typhoon extreme precipitation events over southern China.

This study develops an objective deep-learning-based model for tropical cyclone (TC) intensity estimation. The model’s basic structure is a convolutional neural network (CNN), which is a widely used technology in computer vision tasks. In order to optimize the model’s structure and to improve the feature extraction ability, both residual learning and attention mechanisms are embedded into the model. Five cloud products, including cloud optical thickness, cloud top temperature, cloud top height, cloud effective radius, and cloud type, which are level-2 products from the geostationary satellite Himawari-8, are used as the model training inputs. We sampled the cloud products under the 13 rotational angles of each TC to augment the training dataset. For the independent test data, the model shows improvement, with a relatively low RMSE of 4.06 m/s and a mean absolute error (MAE) of 3.23 m/s, which are comparable to the results seen in previous studies. Various cloud organization patterns, storm whirling patterns, and TC structures from the feature maps are presented to interpret the model training process. An analysis of the overestimated bias and underestimated bias shows that the model’s performance is highly affected by the initial cloud products. Moreover, several controlled experiments using other deep learning architectures demonstrate that our designed model is conducive to estimating TC intensity, thus providing insight into the forecasting of other TC metrics.

Forecasting heavy precipitation accurately is a challenging task for most deep learning (DL)-based models. To address this, we present a novel DL architecture called “multi-scale feature fusion” (MFF) that can forecast precipitation with a lead time of up to 3 h. The MFF model uses convolution kernels with varying sizes to create multi-scale receptive fields. This helps to capture the movement features of precipitation systems, such as their shape, movement direction, and speed. Additionally, the architecture utilizes the mechanism of discrete probability to reduce uncertainties and forecast errors, enabling it to predict heavy precipitation even at longer lead times. For model training, we use 4 years of radar echo data from 2018 to 2021 and 1 year of data from 2022 for model testing. We compare the MFF model with three existing extrapolative models: time series residual convolution (TSRC), optical flow (OF), and UNet. The results show that MFF achieves superior forecast skills with high probability of detection (POD), low false alarm rate (FAR), small mean absolute error (MAE), and high structural similarity index (SSIM). Notably, MFF can predict high-intensity precipitation fields at 3 h lead time, while the other three models cannot. Furthermore, MFF shows improvement in the smoothing effect of the forecast field, as observed from the results of radially averaged power spectral (RAPS). Our future work will focus on incorporating multi-source meteorological variables, making structural adjustments to the network, and combining them with numerical models to further improve the forecast skills of heavy precipitations at longer lead times.

Currently, most deep learning (DL)-based models for precipitation forecasting face two conspicuous issues: the smoothing effect in the precipitation field and the degenerate effect of forecasting precipitation intensity. Therefore, this study proposes “time series residual convolution (TSRC)”, a DL-based convolutional neural network for precipitation nowcasting over China with a lead time of 3 h. The core idea of TSRC is it compensates the current local cues with previous local cues during convolution processes, so more contextual information and less uncertain features would remain in deep networks. We use four years’ radar echo reflectivity data from 2017 to 2020 for model training and one year’s data from 2021 for model testing and compare it with two commonly used nowcasting models: optical flow model (OF) and UNet. Results show that TSRC obtains better forecasting performances than OF and UNet with a relatively high probability of detection (POD), low false alarm rate (FAR), small mean absolute error (MAE) and high structural similarity index (SSIM), especially at longer lead times. Meanwhile, the results of two case studies suggest that TSRC still introduces smoothing effects and slightly outperforms UNet at longer lead times. The most considerable result is that our model can forecast high-intensity radar echoes even for typhoon rainfall systems, suggesting that the degenerate effect of forecasting precipitation intensity can be improved by our model. Future works will focus on the combination of multi-source data and the design of the model’s architecture to gain further improvements in precipitation short-term forecasting.