Convcast: An embedded convolutional LSTM based architecture for precipitation nowcasting using satellite data

Abstract

Nowcasting of precipitation is a difficult spatiotemporal task because of the non-uniform characterization of meteorological structures over time. Recently, convolutional LSTM has been shown to be successful in solving various complex spatiotemporal based problems. In this research, we propose a novel precipitation nowcasting architecture 'Convcast' to predict various short-term precipitation events using satellite data. We train Convcast with ten consecutive NASA's IMERG precipitation data sets each at intervals of 30 minutes. We use the trained neural network model to predict the eleventh precipitation data of the corresponding ten precipitation sequence. Subsequently, the predicted precipitation data are used iteratively for precipitation nowcasting of up to 150 minutes lead time. Convcast achieves an overall accuracy of 0.93 with an RMSE of 0.805 mm/h for 30 minutes lead time, and an overall accuracy of 0.87 with an RMSE of 1.389 mm/h for 150 minutes lead time. Experiments on the test dataset demonstrate that Convcast consistently outperforms other state-of-theart optical flow based nowcasting algorithms. Results from this research can be used for nowcasting of weather events from satellite data as well as for future on-board processing of precipitation data.

Assessing the Potential of Deep Learning for Emulating Cloud Superparameterization in Climate Models With Real‐Geography Boundary Conditions

“…As in Kumar et al. (2020), we show that augmenting our DNN with a positive constraint could reduce the errors in land precipitation emulation (Figure 9b vs. Figure 9d).…”

Enhancing the Encoding-Forecasting Model for Precipitation Nowcasting by Putting High Emphasis on the Latest Data of the Time Step

“…Next, the false alarm rate (FAR), possibility of detection (POD), critical success index (CSI), and the Heidke skill score (HSS) [33] were obtained using the following formulas based on the confusion matrix presented in Table 3. A method of measuring prediction performance according to the level of rainfall rate based on the confusion matrix, which is used in a classification problem, is a common method in the meteorological field [16,18,20,21,24,25,27]. After predicting rainfall through the model, we generate five confusion matrices based on the five thresholds (i.e., the rainfall rate of 0.5, 2.0, 5.0, 10.0, and 30.0 mm/h).…”

“…The second method predicts the output channel of the CNN after changing the time of the RNN to the channel of the CNN [21]. U-Net architecture is mainly used in the CNN-based methods [19,[22][23][24], and stacked convolutional RNN architecture is mainly used in the RNN-based methods [20,21,25,26]. Shi et al [20,21] conducted experiments comprehensively on various methods such as optical flow, 2D CNN, 3D CNN, RNN, and a method of combining CNN and RNN.…”

Nowcasting thunderstorm hazards using machine learning: the impact of data sources on performance

“…ML has been popular for nowcasting precipitation (e.g. Shi et al, 2015Shi et al, , 2017Foresti et al, 2019;Ayzel et al, 2020;Kumar et al, 2020;Franch et al, 2020), and has also been used to develop nowcasting methods for lightning (Mostajabi et al, 2019;Zhou et al, 2020), hail (Czernecki et al, 2019;Huang et al, 2019) and windstorms (Sprenger et al, 2017;Lagerquist et al, 2017Lagerquist et al, , 2020. However, studies so far have typically used only one data source, though in some cases several are utilized.…”