SF-CNN: Signal Filtering Convolutional Neural Network for Precipitation Intensity Estimation

Precipitation intensity estimation is a critical issue in the analysis of weather conditions. Most existing approaches focus on building complex models to extract rain streaks. However, an efficient approach to estimate the precipitation intensity from surveillance cameras is still challenging. This study proposes a convolutional neural network known as the signal filtering convolutional neural network (SF-CNN) to handle precipitation intensity using surveillance-based images. The SF-CNN has two main blocks, the signal filtering block (SF block) and the gradually decreasing dimension block (GDD block), to extract features for the precipitation intensity estimation. The SF block with the filtering operation is constructed in different parts of the SF-CNN to remove the noise from the features containing rain streak information. The GDD block continuously takes the pair of the convolutional operation with the activation function to reduce the dimension of features. Our main contributions are (1) an SF block considering the signal filtering process and effectively removing the useless signals and (2) a procedure of gradually decreasing the dimension of the feature able to learn and reserve the information of features. Experiments on the self-collected dataset, consisting of 9394 raining images with six precipitation intensity levels, demonstrate the proposed approach’s effectiveness against the popular convolutional neural networks. To the best of our knowledge, the self-collected dataset is the largest dataset for monitoring infrared images of precipitation intensity.

Ground validation of GPM IMERG-F precipitation products with the point rain gauge records on the extreme rainfall over a mountainous area of Sumatra Island

Accurate satellite precipitation estimates over areas of complex topography are still challenging, while such accuracy is of importance to the adoption of satellite data for hydrological applications. This study evaluated the ability of Integrated Multi-satellitE Retrievals for GPM -Final (IMERG) V06 product to observe the extreme rainfall over a mountainous area of Sumatra Island. Fifteen years of optical rain gauge (ORG) observation at Kototabang, West Sumatra, Indonesia (100.32°E, 0.20°S, 865 m above sea level), were used as reference surface measurement. The performance of IMERG-F was evaluated using 13 extreme rain indexes formulated by the Expert Team on Climate Change Detection and Indices (ETCCDI). The IMERG-F overestimated the values of all precipitation amount-based indices (PRCPTOT, R85P, R95P, and R99P), three precipitation frequency-based indices (R1mm, R10mm, R20mm), one precipitation duration-based indices (CWD), and one precipitation intensity-based indices (RX5day). Furthermore, the IMERG-F underestimated the values of precipitation frequency-based indices (R50mm), one precipitation duration-based indices (CDD), one precipitation intensity-based indices (SDII). In terms of correlation, only five indexes have a correlation coefficient (R) > 0.5, consistent with Kling–Gupta Efficiency (KGE) value. These results confirm the need to improve the accuracy of the IMERG-F data in mountainous areas.

Simulation and Evaluation of Statistical Downscaling of Regional Daily Precipitation over North China Based on Self-Organizing Maps

A statistical downscaling method based on Self-Organizing Maps (SOM), of which the SOM Precipitation Statistical Downscaling Method (SOM-SD) is named, has received increasing attention. Herein, its applicability of downscaling daily precipitation over North China is evaluated. Six indices (total season precipitation, daily precipitation intensity, mean number of precipitation days, percentage of rainfall from events beyond the 95th percentile value of overall precipitation, maximum consecutive wet days, and maximum consecutive dry days) are selected, which represent the statistics of daily precipitation with regards to both precipitation amount and frequency, as well as extreme event. The large-scale predictors were extracted from the National Center for Environmental Prediction/National Center for Atmospheric Research (NCEP/NCAR) daily reanalysis data, while the prediction was the high resolution gridded daily observed precipitation. The results show that the method can establish certain conditional transformation relationships between large-scale atmospheric circulation and local-scale surface precipitation in a relatively simple way. This method exhibited a high skill in reproducing the climatologic statistical properties of the observed precipitation. The simulated daily precipitation probability distribution characteristics can be well matched with the observations. The values of Brier scores are between 0 and 1.5 × 10−4 and the significance scores are between 0.8 and 1 for all stations. The SOM-SD method, which is evaluated with the six selected indicators, shows a strong simulation capability. The deviations of the simulated daily precipitation are as follows: Total season precipitation (−7.4%), daily precipitation intensity (−11.6%), mean number of rainy days (−3.1 days), percentage of rainfall from events beyond the 95th percentile value of overall precipitation (+3.4%), maximum consecutive wet days (−1.1 days), and maximum consecutive dry days (+3.5 days). In addition, the frequency difference of wet-dry nodes is defined in the evaluation. It is confirmed that there was a significant positive correlation between frequency difference and precipitation. The findings of this paper imply that the SOM-SD method has a good ability to simulate the probability distribution of daily precipitation, especially the tail of the probability distribution curve. It is more capable of simulating extreme precipitation fields. Furthermore, it can provide some guidance for future climate projections over North China.

Extreme precipitation on consecutive days occurs more often in a warming climate

Extreme precipitation occurring on consecutive days may substantially increase the risk of related impacts, but changes in such events have not been studied at a global scale. Here we use a unique global dataset based on in situ observations and multi-model historical and future simulations to analyse the changes in the frequency of extreme precipitation on consecutive days (EPCD). We further disentangle the relative contributions of variations in precipitation intensity and temporal correlation of extreme precipitation, to understand the processes that drive the changes in EPCD. Observations and climate model simulations show that the frequency of EPCD is increasing in most land regions, in particular in North America, Europe and the Northern Hemisphere high latitudes. These increases are primarily a consequence of increasing precipitation intensity, but changes in the temporal correlation of extreme precipitation regionally amplify or reduce the effects of intensity changes. Changes are larger in simulations with a stronger warming signal, suggesting that further increases in EPCD are expected for the future under continued climate warming.

A Novel Multi-Input Multi-Output Recurrent Neural Network Based on Multimodal Fusion and Spatiotemporal Prediction for 0–4 h Precipitation Nowcasting

Multi-source meteorological data can reflect the development process of single meteorological elements from different angles. Making full use of multi-source meteorological data is an effective method to improve the performance of weather nowcasting. For precipitation nowcasting, this paper proposes a novel multi-input multi-output recurrent neural network model based on multimodal fusion and spatiotemporal prediction, named MFSP-Net. It uses precipitation grid data, radar echo data, and reanalysis data as input data and simultaneously realizes 0–4 h precipitation amount nowcasting and precipitation intensity nowcasting. MFSP-Net can perform the spatiotemporal-scale fusion of the three sources of input data while retaining the spatiotemporal information flow of them. The multi-task learning strategy is used to train the network. We conduct experiments on the dataset of Southeast China, and the results show that MFSP-Net comprehensively improves the performance of the nowcasting of precipitation amounts. For precipitation intensity nowcasting, MFSP-Net has obvious advantages in heavy precipitation nowcasting and the middle and late stages of nowcasting.

Study on applicability of remote sensing precipitation products in hilly-plain-wetland complex area of north China

For the hilly-plain-wetland complex ecosystem in the cold region of Northeast China, in order to solve the problems which include the scarcity of surface rainfall stations and the inability to provide accurate surface precipitation for hydrological process simulation, based on the observed precipitation of rainfall stations, three remote sensing precipitation products are taken as objects of evaluation. They include TRMM(Tropical Rainfall Measuring Mission) 3B42V7,3B42RT and CHIRPS (Climate Hazards Group InfraRed Precipitation with Station data,CHIP). In this paper, the observation data of rainfall stations and IDWP precipitation data interpolated by IDW(Inverse Distance Weighted) are used as true value of precision comparison, and the detection accuracy of remote sensing precipitation products from 2001 to 2010 is evaluated on the time scale (day, month and quarter) and spatial scale in Naoli River Basin.The results of the study indicated that 3B42V7 and CHIP have a high detection accuracy for precipitation, and their CC(correlation coefficient) values are 0.47 and 0.49 respectively in daily time scale. The accuracy of their observationfor monthly precipitation is better than that of daily precipitation, and the CC are 0.85 and 0.87 respectively. The multi-year average precipitation at different grid positions in the basin is overestimated by 3B42RT, and its evaluation results are poor at different time scales.For the precipitation intensity range of (0,20], the observed results of 3B42V7 and rainfall station are close to each other. For the precipitation intensity ranges of (0,1) and (50,+∞), 3B42RT and CHIP have overestimated or underestimated the precipitation in different degrees. Based on the above analysis results, 3B42RT can be considered as data that can detect whether precipitation occurs on different spatial positions in the basin. 3B42V7 and CHIP can be applied to flood forecasting and non-point source pollution control in cold regions.

Higher temperature enhances spatio-temporal concentration of rainfall

The relationship between extreme precipitation intensity and temperature has been comprehensively studied over different regions worldwide. However, the effect of temperature on the spatio-temporal organization of precipitation, which can have a significant impact on precipitation intensity, has not been adequately studied or understood. In this study, we propose a novel approach to quantifying the spatial and temporal concentration of precipitation at the event level and study how the concentration varies with temperature. The results based on rain gauge data from 843 stations in the Ganzhou county, a humid region in south China, show that rain events tend to be more concentrated both temporally and spatially at higher temperature, and this increase in concentration qualitatively holds for events of different precipitation amounts and durations. The effects of temperature on precipitation organization in space and in time differ at high temperatures. The temporal concentration increases with temperature up to a threshold (approximately 24°C) beyond which it plateaus, whereas the spatial concentration keeps rising with temperature. More concentrated precipitation, in addition to a projected increase of extreme precipitation, would intensify flooding in a warming world, causing more detrimental effects.

Determining the Precipitation Intensity Threshold of Debris Flood Occurrence

In this chapter, the precipitation threshold at which debris floods occur was evaluated experimentally, and the factors that influence debris flood occurrence, including the bed slope, sediment layer thickness, sediment grain size, length of alluvial flow direction, precipitation intensity, and time of debris flood occurrence, were examined. The impacts of these factors on debris flood initiation were investigated through dimensional analysis. Then, a method was developed to estimate the precipitation intensity threshold based on a set of laboratory tests. Furthermore, different methods for determining the precipitation intensity threshold at which debris floods are initiated were assessed and discussed. The results of the experiments showed that the effect of the sediment layer thickness on debris flood occurrence can be ignored. Moreover, by independently evaluating the effect of each factor on debris flood occurrence, it was found that the sediment length and average diameter of sediments are influential to debris flood initiation. The results of this research provide a better understanding of debris flood mechanisms and occurrence thresholds of debris floods and can be employed to prepare a forecasting model.