“…This is likely due to two factors: (a) convLSTM layers have a more complex logic so require more expertise to tune. Environmental science examples: ConvLSTM is used for example for precipitation forecasting (Shi et al, 2015;Kim et al, 2017;Wang and Hong, 2018;Akbari Asanjan, 2019;Ehsani et al, 2022) and hurricane forecasting Udumulla, 2020).…”
Section: Environmental Data Science E31-13mentioning
confidence: 99%
“…Environmental science examples: ConvLSTM is used for example for precipitation forecasting (Shi et al, 2015; Kim et al, 2017; Wang and Hong, 2018; Akbari Asanjan, 2019; Ehsani et al, 2022) and hurricane forecasting (Kim et al, 2019; Udumulla, 2020).…”
Section: Convolutional Neural Network To Analyze Image Sequencesmentioning
Atmospheric processes involve both space and time. Thus, humans looking at atmospheric imagery can often spot important signals in an animated loop of an image sequence not apparent in an individual (static) image. Utilizing such signals with automated algorithms requires the ability to identify complex spatiotemporal patterns in image sequences. That is a very challenging task due to the endless possibilities of patterns in both space and time. Here, we review different concepts and techniques that are useful to extract spatiotemporal signals from meteorological image sequences to expand the effectiveness of AI algorithms for classification and prediction tasks. We first present two applications that motivate the need for these approaches in meteorology, namely the detection of convection from satellite imagery and solar forecasting. Then we provide an overview of concepts and techniques that are helpful for the interpretation of meteorological image sequences, such as (a) feature engineering methods using (i) meteorological knowledge, (ii) classic image processing, (iii) harmonic analysis, and (iv) topological data analysis; (b) ways to use convolutional neural networks for this purpose with emphasis on discussing different convolution filters (2D/3D/LSTM-convolution); and (c) a brief survey of several other concepts, including the concept of “attention” in neural networks and its utility for the interpretation of image sequences and strategies from self-supervised and transfer learning to reduce the need for large labeled datasets. We hope that presenting an overview of these tools—many of which are not new but underutilized in this context—will accelerate progress in this area.
“…This is likely due to two factors: (a) convLSTM layers have a more complex logic so require more expertise to tune. Environmental science examples: ConvLSTM is used for example for precipitation forecasting (Shi et al, 2015;Kim et al, 2017;Wang and Hong, 2018;Akbari Asanjan, 2019;Ehsani et al, 2022) and hurricane forecasting Udumulla, 2020).…”
Section: Environmental Data Science E31-13mentioning
confidence: 99%
“…Environmental science examples: ConvLSTM is used for example for precipitation forecasting (Shi et al, 2015; Kim et al, 2017; Wang and Hong, 2018; Akbari Asanjan, 2019; Ehsani et al, 2022) and hurricane forecasting (Kim et al, 2019; Udumulla, 2020).…”
Section: Convolutional Neural Network To Analyze Image Sequencesmentioning
Atmospheric processes involve both space and time. Thus, humans looking at atmospheric imagery can often spot important signals in an animated loop of an image sequence not apparent in an individual (static) image. Utilizing such signals with automated algorithms requires the ability to identify complex spatiotemporal patterns in image sequences. That is a very challenging task due to the endless possibilities of patterns in both space and time. Here, we review different concepts and techniques that are useful to extract spatiotemporal signals from meteorological image sequences to expand the effectiveness of AI algorithms for classification and prediction tasks. We first present two applications that motivate the need for these approaches in meteorology, namely the detection of convection from satellite imagery and solar forecasting. Then we provide an overview of concepts and techniques that are helpful for the interpretation of meteorological image sequences, such as (a) feature engineering methods using (i) meteorological knowledge, (ii) classic image processing, (iii) harmonic analysis, and (iv) topological data analysis; (b) ways to use convolutional neural networks for this purpose with emphasis on discussing different convolution filters (2D/3D/LSTM-convolution); and (c) a brief survey of several other concepts, including the concept of “attention” in neural networks and its utility for the interpretation of image sequences and strategies from self-supervised and transfer learning to reduce the need for large labeled datasets. We hope that presenting an overview of these tools—many of which are not new but underutilized in this context—will accelerate progress in this area.
“…The incorporation of deep learning methodologies into rainfall nowcasting represents a substantial advancement in the discipline (Agrawal et al., 2019; Ayzel et al., 2020; Bonnet et al., 2020; Cao et al., 2019; Chen et al., 2020; Choi & Kim, 2021; Ehsani et al., 2021; Fang et al., 2021; Fernández & Mehrkanoon, 2021; Han et al., 2021; Kumar et al., 2020; Luo et al., 2020; Luo, Li, et al., 2022; Luo, Zhao, et al., 2022; Ravuri et al., 2021; Shi et al., 2015; Shi et al., 2017; Tian et al., 2019; Tran & Song, 2019; Trebing et al., 2021; Tuyen et al., 2022; Yan et al., 2020; Yang & Mehrkanoon, 2022). These techniques frame precipitation nowcasting as a spatiotemporal sequence prediction problem.…”
Precise and timely rainfall nowcasting plays a critical role in ensuring public safety amid disasters triggered by heavy precipitation. While deep‐learning models have exhibited superior performance over traditional nowcasting methods in recent years, their efficacy is still hampered by limited forecasting skill, insufficient training data, and escalating blurriness in forecasts. To address these challenges, we present the Synthetic‐data Task‐segmented Generative Model (STGM), an innovative physical‐dynamic‐driven heavy rainfall nowcasting model. The STGM encompasses three key components: the Long Video Generation (LVG) model generating synthetic radar data from observed radar images and data provided by the Weather Research and Forecasting (WRF) model, MaskPredNet predicting the spatial coverage of various rainfall intensities, and SPADE determining rainfall intensity based on the coverage provided by MaskPredNet. The STGM has demonstrated promising skill for precipitation forecasts for up to six hours, and significantly reduce the blurriness of predicted images, thus showcasing advances in rainfall nowcasting.
“…Recently, deep learning (DL) methods have made encouraging progresses in atmosphere-ocean sciences [27][28][29][30][31][32][33]. It allows a model to be fed with raw data as predictors without detailed feature extraction and transformation [34].…”
As most global climate models suffer from large biases in simulating/predicting summer precipitation over China, it is of great importance to develop suitable bias-correction methods. This study proposes two pathways of bias-correction with deep learning (DL) models incorporated. One is the deterministic pathway (DP), in which the bias correction is directly applied to the precipitation forecasts. The other one, namely the probability pathway (PP), corrects the forecasted precipitation anomalies using a conditional probability method before being added to the observational climatology. These two pathways have been applied to correct the precipitation forecasts based on a global climate model prediction system NUIST-CFS1.0. The application of DL models in the both pathways yield higher resolution of corrected predictions than the uncorrected ones. Both pathways improve summer precipitation predictions at 4-month lead. Moreover, the DP correction shows a better performance in predicting extreme precipitation, while the PP is proficient in correcting the spatial pattern of precipitation anomalies over China. The present results highlight the importance of the application of appropriate correction strategy for different prediction purposes.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.